Re: [R] group by-like statement for 2-row matrix
Luc, Eik, Jorge, thanks to all of you! Nice to see how many different solutions there are for the same problem. :) Best Henning Am 07.04.2009 um 22:25 schrieb Jorge Ivan Velez: > > Dear Hans, > > Try also: > > x <- structure(c(2, 1, 2, 1, 3, 2, 4, 1, 4, 3, 4, 3, 5, 2, 5, 1, 6, > 1), .Dim = c(2L, 9L)) > tapply(x[2,],x[1,],sum) > #2 3 4 5 6 > #2 2 7 3 1 > > HTH, > > Jorge > > > On Tue, Apr 7, 2009 at 11:06 AM, Hans-Henning Gabriel > > wrote: > Hi, > > my problem is as follows: > > I have a matrix of two rows like this: > > 2 2 3 4 4 4 5 5 6 > 1 1 2 1 3 3 2 1 1 > > Can I apply something like "group by" in sql? What I want to achieve > is the some of second row for each unique entry of first row: > > 2 -> 2 (=1+1) > 3 -> 2 > 4 -> 7 (=1+3+3) > 5 -> 3 (=2+1) > 6 -> 1 > > Thanks!! > Henning > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Re ading Excel 5.0 files with RODBC?
Thanks for reply David, but the problem is that the place i get those files from saves them in the old Excel format (Excel 5.0), not Excel 97-2003, so neither of the packages (RODBC, xlsReadWrite) can read it. >Your question is a bit confusing. This is the current state of play >regarding reading and writing Excel files with xlsReadWrite and RODBC. > >xlsReadWrite will read and write from Excel 97-2003 files (.xls files) >including reading from named sheets, but only writing a single sheet >workbook. Regards, Yuri -- View this message in context: http://www.nabble.com/Reading-Excel-5.0-files-with-RODBC--tp22929781p22944550.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] References for elliptical2d(fCopulae)
Dear R-mailing list, I am using the fCopulae package. Where can I find the relevant references for functions such as elliptical2d( ) ? I am particularly interested in references pointing out how the correlation coefficients can be incorporated in such density functions. Thanks in advance Caspar Hallmann [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Student
Hi, I have problem. In the function below (test and test2) i want the function test not to print the variable data but i want the function test2 to use the variable test$data. This is the creation of the variable data: > matrice=c(1:10) > matrice=matrix(matrice,nrow=5,ncol=2) This is the function test: > test=function(data){ + return(list(x=5,data=data)) + } This is the function test2: > test2=function(list){ + bodri=list$data + bodri[1,2]=bodri[2,2]+1 + return(bodri) + } Below there are the result: > uno=test(matrice) > due=test2(uno) > uno $x [1] 5 $data [,1] [,2] [1,]16 [2,]27 [3,]38 [4,]49 [5,]5 10 > due [,1] [,2] [1,]18 [2,]27 [3,]38 [4,]49 [5,]5 10 What i want is: > uno=test(matrice) > due=test2(uno) > uno $x [1] 5 > due [,1] [,2] [1,]18 [2,]27 [3,]38 [4,]49 [5,]5 10 Can you help me please? Thank you. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] help loading an svm model from file
Hi R-experts, I'm using the svm implentation available in the library 'e1071', to train a classifier using cross validation (...) obj = tune.svm(truth ~ ., data=examples,cost=2^(grid.cost),gamma=2^(grid.gamma),cross=k) and to save the best model into a file as ?write.svm suggests: write.svm(obj$best.model, svm.file = "Rdata.svm",scale.file = "Rdata_svm.scale") My problem is when I want to read/load the Rdata.svm which contains the R object corresponding to the svm model. I didn't find the way to recover the data saved in Rdata.svm using write.svm. how can I do that? Thanks, John Disfruta antes que nadie del nuevo Windows Live Messenger _ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] asymptotic convergence intervall for poisson regression
Hello! The following code is an implementation of a Poisson regression. It generates some data-samples and computes the beta values with the negative log likelihood function. Now, my task is to compute the asymptotic convergence intervalls for the values of beta but I dont know how to implement this function - this topic is not in my lecture notes. I hope someone can help me. > library(Bhat) > # generate new data > dose <- c(rep(0,50), rep(1,50), rep(5,50), rep(10,50)) > data <- cbind(dose, rpois(200,2*(1+(10-dose) *.5*(1-(10-dose)*0.05 > data > lambda <- function(dose) > { > 2*(1+(10 - dose) * .5 * (1-(10-dose)*0.05)) > } > plot(c(0:10),lambda(c(0:10))) > # estimated count of fits - dose 0:10 > plot (data[,1] + rnorm(200, mean=0, sd=0.15), data[,2]) > # Likelihood - function > negloglike <- function(beta) > { > ds <- data[,1] > x <- data[,2] > lambda <- beta[1] * (1 + ( 10 - ds ) * beta[2] * (1 - (10 - ds) * beta[3])) > return(sum(lambda - x * log(lambda))) > } > beta <- list(label = c("beta1","beta2","beta3"), est=c(2.5,0.5,0.1), low=c(1,0,0),upp=c(3,2,2)) > result <- dfp(beta,f=negloglike) > result Kind regards -- View this message in context: http://www.nabble.com/asymptotic-convergence-intervall-for-poisson-regression-tp22944357p22944357.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] [R-pkgs] tcltk - control message while the program is running
Hi, We are trying to write control messages (errors, warnings, ...) to a tktable while the program is running. Until now we only managed to write all the messages at once at the end of execution. How can we refresh the window in order to show the tclArray content? ... options(show.error.messages = FALSE, warn=1) base <- try(suppressWarnings(DesenhoRegisto(regiao,ano,trimestre))) if (class(base) == "try-error") { msg <- paste("Desenho de registo:",as.character(base),sep=" ") msg <- sub("\n","",msg) return(list(tabela=NULL,msg=msg)) } msg <- "Desenho de Registo...Ok" print(msg) tclArray1[[1,0]]<<-as.tclObj(msg, drop=TRUE) tkactivate(table1, "1,1") tksee(table1, "1,1") message("testing") aloj <- data.frame(base[2]) base <- data.frame(base[1]) base <- try(suppressWarnings(VariaveisAuxiliares(base,ano,trimestre))) if (class(base) == "try-error") { msg <- paste("Variáveis auxiliares:",as.character(base),sep=" ") msg <- sub("\n","",msg) return(list(tabela=NULL,msg=msg)) } msg <- "Variáveis Auxiliares...Ok" tkinsert(table1,"rows","end",1) tclArray1[[2,0]]<<-as.tclObj(msg, drop=TRUE) tkactivate(table1, "2,1") tksee(table1, "2,1") ... Thank you all, SRPD _ [[alternative HTML version deleted]] ___ R-packages mailing list r-packa...@r-project.org __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Re ading Excel 5.0 files with RODBC?
On Wed, 8 Apr 2009, Yuri Volchik wrote: Thanks for reply David, but the problem is that the place i get those files from saves them in the old Excel format (Excel 5.0), not Excel 97-2003, so neither of the packages (RODBC, xlsReadWrite) can read it. RODBC supports any format for which an ODBC driver is installed and Microsoft have long supplied ones for Excel 3.0 and later (up to Excel 2007 at present). -- Brian D. Ripley, rip...@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Minimum Spanning Tree
On Tue, Apr 7, 2009 at 11:06 PM, jpearl01 wrote: > > There was an error in the file... an extraneous comma. That's taken care of. > however, my tree prints out an image that doesn't seem like a mst. Attached > is the csv file I used... Well, it looks definitely a tree to me. > http://www.nabble.com/file/p22938299/sp_matrix.csv sp_matrix.csv > > I'd like it to look something like the image file also attached... > http://www.nabble.com/file/p22938299/2006-08-27_MST.png 2006-08-27_MST.png > > Is there a different layout that would accomplish that? Or if not that > exactly, one that would help make the results a little clearer? I am not sure what you mean. Of course you can plot it using different layouts, e.g. with layout.reingold.tilford (after choosing the root vertex in some way) and then it looks like a usual tree plot, but why would that be any better? Unless there is some external information about the graph (e.g. spatial positions of the nodes, or a distinguished root vertex), the layout on the image is just as good as the others. Gabor > Thanks for all the help! > ~josh > -- > View this message in context: > http://www.nabble.com/Minimum-Spanning-Tree-tp22934813p22938299.html > Sent from the R help mailing list archive at Nabble.com. > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Gabor Csardi UNIL DGM __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Annotation rgl plot
Thanks for a great package Duncan, one of the most amazing things i saw for R platform. In the meanwhile i can do with you suggestion: par3d(ignoreExtent=TRUE); -- View this message in context: http://www.nabble.com/Annotation-rgl-plot-tp22930428p22945459.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Student
On Wed, Apr 08, 2009 at 10:02:10AM +0200, alberto cassese wrote: > Hi, > I have problem. In the function below (test and test2) i want the function > test not to print the variable data but i want the function test2 to use the > variable test$data. > > This is the creation of the variable data: > > > matrice=c(1:10) > > matrice=matrix(matrice,nrow=5,ncol=2) > > This is the function test: > > > test=function(data){ > + return(list(x=5,data=data)) > + } > > This is the function test2: > > > test2=function(list){ > + bodri=list$data > + bodri[1,2]=bodri[2,2]+1 > + return(bodri) > + } > > Below there are the result: > > > uno=test(matrice) > > due=test2(uno) > > uno > $x > [1] 5 > > $data > [,1] [,2] > [1,]16 > [2,]27 > [3,]38 > [4,]49 > [5,]5 10 > > > due > [,1] [,2] > [1,]18 > [2,]27 > [3,]38 > [4,]49 > [5,]5 10 > > > What i want is: > > > uno=test(matrice) > > due=test2(uno) > > uno > $x > [1] 5 x is a variable, 5 is variable data and you don't want variable data printed? > > due > [,1] [,2] > [1,]18 > [2,]27 > [3,]38 > [4,]49 > [5,]5 10 > Use uno[1], either directly or by creating a third variable from uno[1] > one.and.a.half <- uno[1] > one.and.a.half $x [1] 5 Or, if you *really* want what that printed output from test(matrice), create a class for your list-object, and add a special print method, that will only print the first item of the list. -- Hans Ekbrand (http://sociologi.cjb.net) A. Because it breaks the logical sequence of discussion Q. Why is top posting bad? signature.asc Description: Digital signature __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Tinn-R pdf()
Dear R and Tinn-R users, i recently switched to Tinn-R and sending code to R works fine (R 2.8.1, Tinn-R 2.2.0.2, OS Windows XP). However, i encountered a problem when trying to send plots to pdf files like this: library(lattice) pdf("plot1.pdf") PLOT<-(xyplot, ...) PLOT dev.off() The file "plot1.pdf" is created, but it is empty. If i paste the code above directly into the R console and run it, the file "plot1.pdf" is created and in this case contains "PLOT". I guess that some settings in Tinn-R are wrong, but i have no idea which. Maybe someone has a suggestion? Thanks, Henning -- [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Clear all with/without function()
why rm(list=ls(all=TRUE),envir=globalenv()) is ok but ca<-function() rm(list=ls(all=TRUE),envir=globalenv()) ca() does not work? Thank you! v -- View this message in context: http://www.nabble.com/Clear-all-with-without-function%28%29-tp22947021p22947021.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] read.spss, locale and encodings
I must be missing something obvious here: According to the help page for read.spss, the reencode option is only active when R is run under a UTF-8 locale. read.spss can only import the SPSS file when run under a iso88591(5) locale, under a UTF-8 locale I get: Error in read.spss("wo.sav") : error reading system-file header In addition: Warning message: In read.spss("wo.sav") : wo.sav: position 143: Variable name begins with invalid character This is under Debian GNU/Linux, the stable release. foreign is version 8.27 -- Hans Ekbrand (http://sociologi.cjb.net) GPG Fingerprint: 1408 C8D5 1E7D 4C9C C27E 014F 7C2C 872A 7050 614E signature.asc Description: Digital signature __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Clear all with/without function()
Taraxacum88 wrote: why rm(list=ls(all=TRUE),envir=globalenv()) is ok but ca<-function() rm(list=ls(all=TRUE),envir=globalenv()) ca() does not work? ls(all=TRUE) from within your function lists all local variables, not globals. You need envir=globalenv() there too. Duncan Murdoch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Tinn-R pdf()
Dear Henning, You need to print() lattice plots when using a device: library(lattice) pdf("plot1.pdf") PLOT<-(xyplot, ...) print(PLOT) dev.off() So this is not due to TINN-R. HTH, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 thierry.onkel...@inbo.be www.inbo.be To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of. ~ Sir Ronald Aylmer Fisher The plural of anecdote is not data. ~ Roger Brinner The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. ~ John Tukey -Oorspronkelijk bericht- Van: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] Namens Henning Wildhagen Verzonden: woensdag 8 april 2009 11:25 Aan: r-help@r-project.org Onderwerp: [R] Tinn-R pdf() Dear R and Tinn-R users, i recently switched to Tinn-R and sending code to R works fine (R 2.8.1, Tinn-R 2.2.0.2, OS Windows XP). However, i encountered a problem when trying to send plots to pdf files like this: library(lattice) pdf("plot1.pdf") PLOT<-(xyplot, ...) PLOT dev.off() The file "plot1.pdf" is created, but it is empty. If i paste the code above directly into the R console and run it, the file "plot1.pdf" is created and in this case contains "PLOT". I guess that some settings in Tinn-R are wrong, but i have no idea which. Maybe someone has a suggestion? Thanks, Henning -- [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver weer en binden het INBO onder geen enkel beding, zolang dit bericht niet bevestigd is door een geldig ondertekend document. The views expressed in this message and any annex are purely those of the writer and may not be regarded as stating an official position of INBO, as long as the message is not confirmed by a duly signed document. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] change default output size when using Sweave
On 8 Apr 2009, at 11:44, Duncan Murdoch wrote: Mark Heckmann wrote: Dear Duncan, Thanks for the reply. This works, but unfortunately I need a different solution. My script is supposed to run completely automated and the graphics I produce vary in size each time I run the script. But I want the graphics to be fitted to my .pdf output without specifying the height argument manually each time. If your documents are more or less fixed, you could add an extra preprocessor step before running Sweave. For example, create mydoc.in containing <>= ... @ and then use sed or some other text substitution utility (even R, by reading your whole document into an R variable) to substitute the desired size for the string @s...@. It's clunky, but it sounds as though you're interested in producing standardized documents regularly, so it may be acceptable. For that task you might find interest in the brew package by Jeffrey Horner. baptiste Duncan Murdoch That is why I do not want a fixed height as a code chunk argument. Actually I do not know if it is possible to have a variable placed in a code chunk header. I tried the following which does not work: <<>>= size <- 3 @ <>= pushViewport(viewport(height = unit(80, "mm"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport() @ So still I face the problem to have Sweave generate a .pdf graphic that is just as big I want it to be. In the Sweave Docu paragraph A.9 I discovered something I use as a workaround. I produce the .pdf output manually (where I can control the size) and add each graphic to LaTex manually as well. <>= for (i in 1){ file=paste("myfile", i, ".pdf", sep="") pdf(file=file, paper="special", width=6, height=3) pushViewport(viewport(height = unit(5, "inches"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport()dev.off() cat("\\includegraphics{", file, "}\n\n", sep="") } @ I use parenthesis around the code as it prints out something I do not want if no parenthesis are used (I use Windows). I am not too happy with the solution. I would prefer a more straightforward approach to define the size of the output graphic. I wonder if there are some Sweave settings that can be modified. In the Sweave manual (A.11) I found the following to customize the par settings for each figure: options(SweaveHooks=list(fig=function() par(bg="red", fg="blue"))) I wonder if something similar could be done changing the size of the default output device (pdf or eps) for each figure like >pdf.options(height=2) or similar (it seems that this does not work)? I suppose this type of graphic customization is quite a common issue when producing automated customized output/reports using R and Sweave but I haven't found anything concerning this topic yet. So I would be really glad if someone knows a solution. TIA, Mark -Ursprüngliche Nachricht- Von: Duncan Murdoch [mailto:murd...@stats.uwo.ca] Gesendet: Dienstag, 7. April 2009 19:08 An: Mark Heckmann Cc: r-help@r-project.org Betreff: Re: [R] change default output size when using Sweave On 4/7/2009 10:39 AM, Mark Heckmann wrote: I am trying to create a graphic output in Sweave but I do not want it to be standard size. I want the whole graphic to be 80mm of height only, just like the viewport below. <>= pushViewport(viewport(height = unit(80, "mm"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport() @ How can I make the graphic smaller (80mm of height) without having the white margins at the top and bottom? You can specify height and width as arguments to the code chunk or set new defaults with something like \SweaveOpts{width=5,height=3} (I think those measurements need to be given in inches; I don't think there's a way to use an expression like unit(80,"mm").) You can also control the size of the plot in the final document with LaTeX code like \usepackage{graphicx} \setkeys{Gin}{width=3in} and then the LaTeX rules for handling units would apply, so 80mm is fine instead of 3in. Duncan Murdoch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. _ Baptiste Auguié School of Physics University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag __ R-help@r-project.org mailing list
Re: [R] Tinn-R pdf()
Dear Thierry, thanks for your help, with solved the problem, although i don't why, because when using the R editor accessible via the R console i created many many lattice plots with the code i posted, i.e. without the print() command. Anyway, thanks a lot, Henning > > Original-Nachricht > Datum: Wed, 8 Apr 2009 12:29:58 +0200 > Von: "ONKELINX, Thierry" > An: "Henning Wildhagen" , r-help@r-project.org > Betreff: RE: [R] Tinn-R pdf() > > Dear Henning, > > You need to print() lattice plots when using a device: > > library(lattice) > > pdf("plot1.pdf") > PLOT<-(xyplot, ...) > print(PLOT) > dev.off() > > So this is not due to TINN-R. > > HTH, > > Thierry > > > > ir. Thierry Onkelinx > Instituut voor natuur- en bosonderzoek / Research Institute for Nature > and Forest > Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, > methodology and quality assurance > Gaverstraat 4 > 9500 Geraardsbergen > Belgium > tel. + 32 54/436 185 > thierry.onkel...@inbo.be > www.inbo.be > > To call in the statistician after the experiment is done may be no more > than asking him to perform a post-mortem examination: he may be able to > say what the experiment died of. > ~ Sir Ronald Aylmer Fisher > > The plural of anecdote is not data. > ~ Roger Brinner > > The combination of some data and an aching desire for an answer does not > ensure that a reasonable answer can be extracted from a given body of > data. > ~ John Tukey > > -Oorspronkelijk bericht- > Van: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] > Namens Henning Wildhagen > Verzonden: woensdag 8 april 2009 11:25 > Aan: r-help@r-project.org > Onderwerp: [R] Tinn-R pdf() > > Dear R and Tinn-R users, > > i recently switched to Tinn-R and sending code to R works fine (R 2.8.1, > > Tinn-R 2.2.0.2, OS Windows XP). However, i encountered a problem when > trying to send plots to pdf files like this: > > library(lattice) > > pdf("plot1.pdf") > PLOT<-(xyplot, ...) > PLOT > dev.off() > > The file "plot1.pdf" is created, but it is empty. > If i paste the code above directly into the R console and run it, the > file > "plot1.pdf" is created and in this case contains "PLOT". > > I guess that some settings in Tinn-R are wrong, but i have no idea > which. > Maybe someone has a suggestion? > > Thanks, > > Henning > -- > > > [[alternative HTML version deleted]] > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > > Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver > weer > en binden het INBO onder geen enkel beding, zolang dit bericht niet > bevestigd is > door een geldig ondertekend document. The views expressed in this > message > and any annex are purely those of the writer and may not be regarded as > stating > an official position of INBO, as long as the message is not confirmed by > a duly > signed document. > -- [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] factor, as.factor and levels
Dear All, to my surprise as.factor does not accept a levels argument. Maybe I did not read the documentation well enough. See the example below. I wanted to use ch1 as factor in the newdata argument of survfit, so I assumed that I could write as.factor(ch1, levels=ch1), since the order should be kept. But as.factor(ch1, levels=ch1) results in the error: Error in as.factor(ch1, levels = ch1) : unused argument(s) (levels = c("low", "inter", "high")) factor(ch1, levels=ch1) works as I expected. Is it intended that as.factor does not use the levels argument? Thanks, Heinz ch1 <- c('low', 'inter', 'high') factor(ch1) factor(ch1, levels=ch1) as.factor(ch1, levels=ch1) > version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status Patched major 2 minor 8.1 year 2009 month 03 day13 svn rev48132 language R version.string R version 2.8.1 Patched (2009-03-13 r48132) > sessionInfo() R version 2.8.1 Patched (2009-03-13 r48132) i386-pc-mingw32 locale: LC_COLLATE=German_Switzerland.1252;LC_CTYPE=German_Switzerland.1252;LC_MONETARY=German_Switzerland.1252;LC_NUMERIC=C;LC_TIME=German_Switzerland.1252 attached base packages: [1] splines stats graphics grDevices utils datasets methods [8] base other attached packages: [1] survival_2.34-1 car_1.2-12 gmodels_2.14.1 gdata_2.4.2 [5] Hmisc_3.5-2 loaded via a namespace (and not attached): [1] cluster_1.11.12 grid_2.8.1 gtools_2.5.0-1 lattice_0.17-20 [5] MASS_7.2-46 > __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] help with random forest package
Hello, I am a phd student in Bioinformatics and I am using the Random Forest package in order to classify my data, but I have some questions. Is there a function in order to visualize the trees, so as to get the rules? Also, could you please provide me with the code of "randomForest" function, as I would like to see how it works. I was wondering if I can get the classification having the most votes over all the trees in the forest (the final rules that will give me the final classification). Also, is there a possibility to get a vector with the attributes that are being selected for each node during the construction of each tree? I mean, that I would like to know the m
Re: [R] factor, as.factor and levels
Thank you, Jim. I see, the fact that in the documentation you find only "as.factor(x)" means that it does not accept more arguments. Does as.factor have speed advantages over factor, or is there a different cause for it's existence? Heinz At 13:50 08.04.2009, jim holtman wrote: as.factor does not accept levels as an argument. use the first form that you have factor(ch1, levels=ch1) On Wed, Apr 8, 2009 at 7:36 AM, Heinz Tuechler wrote: > Dear All, > > to my surprise as.factor does not accept a levels argument. Maybe I did not > read the documentation well enough. See the example below. I wanted to use > ch1 as factor in the newdata argument of survfit, so I assumed that I could > write as.factor(ch1, levels=ch1), since the order should be kept. > > But as.factor(ch1, levels=ch1) results in the error: > > Error in as.factor(ch1, levels = ch1) : > unused argument(s) (levels = c("low", "inter", "high")) > > factor(ch1, levels=ch1) works as I expected. > Is it intended that as.factor does not use the levels argument? > > Thanks, > > Heinz > > ch1 <- c('low', 'inter', 'high') > factor(ch1) > factor(ch1, levels=ch1) > as.factor(ch1, levels=ch1) > >> version > _ > platform i386-pc-mingw32 > arch i386 > os mingw32 > system i386, mingw32 > status Patched > major 2 > minor 8.1 > year 2009 > month 03 > day13 > svn rev48132 > language R > version.string R version 2.8.1 Patched (2009-03-13 r48132) >> sessionInfo() > R version 2.8.1 Patched (2009-03-13 r48132) > i386-pc-mingw32 > > locale: > LC_COLLATE=German_Switzerland.1252;LC_CTYPE=German_Switzerland.1252;LC_MONETARY=German_Switzerland.1252;LC_NUMERIC=C;LC_TIME=German_Switzerland.1252 > > attached base packages: > [1] splines stats graphics grDevices utils datasets methods > [8] base > > other attached packages: > [1] survival_2.34-1 car_1.2-12 gmodels_2.14.1 gdata_2.4.2 > [5] Hmisc_3.5-2 > > loaded via a namespace (and not attached): > [1] cluster_1.11.12 grid_2.8.1 gtools_2.5.0-1 lattice_0.17-20 > [5] MASS_7.2-46 >> > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Sequences
Hi Melissa, so what you want to calculate is something like S<-cumsum(lambs)-(1:length(lambs))*mean(lambs) since (with x.bar=mean(x)) S_1=x_1-x.bar S_2=S1+(x_2-x.bar)=x_1-x.bar+x_2-x_bar=x_1+x_2-2*x.bar ... S_n=S_(n-1)+x_n-x.bar=sum_i (x_i)-n*x.bar Eik. m.mcquil...@lancaster.ac.uk schrieb: Hi, I am trying to perform a change point analysis. I know there is a function that does this but I am just out of university on an internship so they are trying to make me do it myself using this website : http://www.variation.com/cpa/tech/changepoint.html So if you look for CUMSUM on this you will see what I am trying to do but I'm not having much success! Melissa -- Eik Vettorazzi Institut für Medizinische Biometrie und Epidemiologie Universitätsklinikum Hamburg-Eppendorf Martinistr. 52 20246 Hamburg T ++49/40/42803-8243 F ++49/40/42803-7790 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] change default output size when using Sweave
Mark Heckmann wrote: Dear Duncan, Thanks for the reply. This works, but unfortunately I need a different solution. My script is supposed to run completely automated and the graphics I produce vary in size each time I run the script. But I want the graphics to be fitted to my .pdf output without specifying the height argument manually each time. If your documents are more or less fixed, you could add an extra preprocessor step before running Sweave. For example, create mydoc.in containing <>= ... @ and then use sed or some other text substitution utility (even R, by reading your whole document into an R variable) to substitute the desired size for the string @s...@. It's clunky, but it sounds as though you're interested in producing standardized documents regularly, so it may be acceptable. Duncan Murdoch That is why I do not want a fixed height as a code chunk argument. Actually I do not know if it is possible to have a variable placed in a code chunk header. I tried the following which does not work: <<>>= size <- 3 @ <>= pushViewport(viewport(height = unit(80, "mm"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport() @ So still I face the problem to have Sweave generate a .pdf graphic that is just as big I want it to be. In the Sweave Docu paragraph A.9 I discovered something I use as a workaround. I produce the .pdf output manually (where I can control the size) and add each graphic to LaTex manually as well. <>= for (i in 1){ file=paste("myfile", i, ".pdf", sep="") pdf(file=file, paper="special", width=6, height=3) pushViewport(viewport(height = unit(5, "inches"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport()dev.off() cat("\\includegraphics{", file, "}\n\n", sep="") } @ I use parenthesis around the code as it prints out something I do not want if no parenthesis are used (I use Windows). I am not too happy with the solution. I would prefer a more straightforward approach to define the size of the output graphic. I wonder if there are some Sweave settings that can be modified. In the Sweave manual (A.11) I found the following to customize the par settings for each figure: options(SweaveHooks=list(fig=function() par(bg="red", fg="blue"))) I wonder if something similar could be done changing the size of the default output device (pdf or eps) for each figure like >pdf.options(height=2) or similar (it seems that this does not work)? I suppose this type of graphic customization is quite a common issue when producing automated customized output/reports using R and Sweave but I haven't found anything concerning this topic yet. So I would be really glad if someone knows a solution. TIA, Mark -Ursprüngliche Nachricht- Von: Duncan Murdoch [mailto:murd...@stats.uwo.ca] Gesendet: Dienstag, 7. April 2009 19:08 An: Mark Heckmann Cc: r-help@r-project.org Betreff: Re: [R] change default output size when using Sweave On 4/7/2009 10:39 AM, Mark Heckmann wrote: I am trying to create a graphic output in Sweave but I do not want it to be standard size. I want the whole graphic to be 80mm of height only, just like the viewport below. <>= pushViewport(viewport(height = unit(80, "mm"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport() @ How can I make the graphic smaller (80mm of height) without having the white margins at the top and bottom? You can specify height and width as arguments to the code chunk or set new defaults with something like \SweaveOpts{width=5,height=3} (I think those measurements need to be given in inches; I don't think there's a way to use an expression like unit(80,"mm").) You can also control the size of the plot in the final document with LaTeX code like \usepackage{graphicx} \setkeys{Gin}{width=3in} and then the LaTeX rules for handling units would apply, so 80mm is fine instead of 3in. Duncan Murdoch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Tinn-R pdf()
Hi Henning, thanks for your help, with solved the problem, although i don't why, because when using the R editor accessible via the R console i created many many lattice plots with the code i posted, i.e. without the print() command. At the command line, R objects (including lattice plots which are objects) get auto-printed, i.e. the print method is invoked automatically on these objects. This is not the case when you write to a pdf file. Best, Tobias Original-Nachricht Datum: Wed, 8 Apr 2009 12:29:58 +0200 Von: "ONKELINX, Thierry" An: "Henning Wildhagen" , r-help@r-project.org Betreff: RE: [R] Tinn-R pdf() Dear Henning, You need to print() lattice plots when using a device: library(lattice) pdf("plot1.pdf") PLOT<-(xyplot, ...) print(PLOT) dev.off() So this is not due to TINN-R. HTH, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 thierry.onkel...@inbo.be www.inbo.be To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of. ~ Sir Ronald Aylmer Fisher The plural of anecdote is not data. ~ Roger Brinner The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. ~ John Tukey -Oorspronkelijk bericht- Van: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] Namens Henning Wildhagen Verzonden: woensdag 8 april 2009 11:25 Aan: r-help@r-project.org Onderwerp: [R] Tinn-R pdf() Dear R and Tinn-R users, i recently switched to Tinn-R and sending code to R works fine (R 2.8.1, Tinn-R 2.2.0.2, OS Windows XP). However, i encountered a problem when trying to send plots to pdf files like this: library(lattice) pdf("plot1.pdf") PLOT<-(xyplot, ...) PLOT dev.off() The file "plot1.pdf" is created, but it is empty. If i paste the code above directly into the R console and run it, the file "plot1.pdf" is created and in this case contains "PLOT". I guess that some settings in Tinn-R are wrong, but i have no idea which. Maybe someone has a suggestion? Thanks, Henning -- [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver weer en binden het INBO onder geen enkel beding, zolang dit bericht niet bevestigd is door een geldig ondertekend document. The views expressed in this message and any annex are purely those of the writer and may not be regarded as stating an official position of INBO, as long as the message is not confirmed by a duly signed document. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Convert data frame containing time stamps to time series
I read records using scan: dat<-data.frame(scan(file="KDA.csv",what=list(t="%m/%d/%y %H:%M",f=0,p=0,d=0,o=0,s=0,a=0,l=0,c=0),skip=2,sep=",",nmax=np,flush=TRUE,na.strings=c("I/OTimeout","ArcOff-line"))) which results in: > dat[1:5,] t fp d o sa l c 1 1/21/09 5:01 16151 8.2 76 30 282 1060 53 7 2 1/21/09 5:02 16256 8.3 76 23 282 1059 54 7 3 1/21/09 5:03 16150 8.4 76 26 282 1059 55 7 4 1/21/09 5:04 16150 9.0 76 25 282 1051 57 6 5 1/21/09 5:05 15543 10.4 76 7 282 1024 58 6 I have been unable to find a way to convert this into a time series. I did read the manuals and came across a way to coerce a data frame to a ts object: as.ts() Trouble is I do not know how to keep the timestamps in column t in the data frame above. The t column is not strings. If I do: plot.ts(dat) I can see how the first graphics panel is indeed numbers not text. So I think scan converted the text correctly per the format string I put in. Much more difficult still. The datafiles I have contain invalid data, missing values and other none relevant information. I filter this out using subset which works brilliantly. However, how can I filter using subset and convert to a time series afterwards. Since after subsetting there will be 'holes' i.e. missing records. Can a ts object deal with missing records? If so, how? Just point me to a document. I can and will put in the work to figure it out myself. Thank you! Alex van der Spek __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Howto Plot With Transparent Background
Is there a way to do it in R? Especially generating plot in EPS/PDF format. By transparent I mean clear (not white) background. I want to attached it to dark PPT slides. - Gundala Viswanath __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Clear all with/without function()
clr <- function () rm(list = ls(pos = .GlobalEnv), pos = .GlobalEnv) this works for me HTH, baptiste On 8 Apr 2009, at 10:50, Taraxacum88 wrote: why rm(list=ls(all=TRUE),envir=globalenv()) is ok but ca<-function() rm(list=ls(all=TRUE),envir=globalenv()) ca() does not work? Thank you! v -- View this message in context: http://www.nabble.com/Clear-all-with-without-function%28%29-tp22947021p22947021.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. _ Baptiste Auguié School of Physics University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Null-Hypothesis
Hello R users, I've used the following help two compare two regression line slopes. Wanted to test if they differ significantly: Hi, I've made a research about how to compare two regression line slopes (of y versus x for 2 groups, "group" being a factor ) using R. I knew the method based on the following statement : t = (b1 - b2) / sb1,b2 where b1 and b2 are the two slope coefficients and sb1,b2 the pooled standard error of the slope (b) which can be calculated in R this way: > df1 <- data.frame(x=1:3, y=1:3+rnorm(3)) > df2 <- data.frame(x=1:3, y=1:3+rnorm(3)) > fit1 <- lm(y~x, df1) > s1 <- summary(fit1)$coefficients > fit2 <- lm(y~x, df2) > s2 <- summary(fit2)$coefficients > db <- (s2[2,1]-s1[2,1]) > sd <- sqrt(s2[2,2]^2+s1[2,2]^2) > df <- (fit1$df.residual+fit2$df.residual) > td <- db/sd > 2*pt(-abs(td), df) My value I get by running this test is :[1] 2.305553e-07 Does it mean the two slopes differ significantly, because this value is in the alpha area, so that I have to reject the null- hypothesis and accept the alternative hypothesis? Is the null-hypothesis: slope1=slope2? Thanks for your help, Benedikt -- __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] newbie query: simple crosstabs
On Tue, 2009-04-07 at 16:33 -0400, Donald Braman wrote: > I've been playing around with various table tools, trying to construct a > fairly simple cross-tab. It shouldn't be hard, but for some reason it > turning out to be (for me). > > If I want to see how many men and how many women agree with a agree/disagree > question (coded 1,0), I can do this: > > >attach(mydata) > >mytable <- table(male, q1.bin) # gender and a binary response variable > >prop.table(mytable, 1) # row percentages > q1.bin > male 0 1 >0 0.3988 0.6012 >1 0.2879 0.7121 > > I can repeat that for each of the items I want gender breakdowns for (q2, > q3, q4 ). But what I really want is a table that shows the percentage > answering yes (coded as 1) across many, many binary response items. E.g., > > > male q1.bin q2.bin q3.bin ... >0 0.6012 0.3421 0.9871 ... >1 0.7121 0.6223 0.0198 ... > > I've tried various combinations of apply & cbind, but to no avail. It would > be easy in SPSS crosstabs, but darnit, I want to use R! Donald, The other solutions is using command CrossTable of package gmodels. -- Bernardo Rangel Tura, M.D,MPH,Ph.D National Institute of Cardiology Brazil __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] factor, as.factor and levels
as.factor does not accept levels as an argument. use the first form that you have factor(ch1, levels=ch1) On Wed, Apr 8, 2009 at 7:36 AM, Heinz Tuechler wrote: > Dear All, > > to my surprise as.factor does not accept a levels argument. Maybe I did not > read the documentation well enough. See the example below. I wanted to use > ch1 as factor in the newdata argument of survfit, so I assumed that I could > write as.factor(ch1, levels=ch1), since the order should be kept. > > But as.factor(ch1, levels=ch1) results in the error: > > Error in as.factor(ch1, levels = ch1) : > unused argument(s) (levels = c("low", "inter", "high")) > > factor(ch1, levels=ch1) works as I expected. > Is it intended that as.factor does not use the levels argument? > > Thanks, > > Heinz > > ch1 <- c('low', 'inter', 'high') > factor(ch1) > factor(ch1, levels=ch1) > as.factor(ch1, levels=ch1) > >> version > _ > platform i386-pc-mingw32 > arch i386 > os mingw32 > system i386, mingw32 > status Patched > major 2 > minor 8.1 > year 2009 > month 03 > day 13 > svn rev 48132 > language R > version.string R version 2.8.1 Patched (2009-03-13 r48132) >> sessionInfo() > R version 2.8.1 Patched (2009-03-13 r48132) > i386-pc-mingw32 > > locale: > LC_COLLATE=German_Switzerland.1252;LC_CTYPE=German_Switzerland.1252;LC_MONETARY=German_Switzerland.1252;LC_NUMERIC=C;LC_TIME=German_Switzerland.1252 > > attached base packages: > [1] splines stats graphics grDevices utils datasets methods > [8] base > > other attached packages: > [1] survival_2.34-1 car_1.2-12 gmodels_2.14.1 gdata_2.4.2 > [5] Hmisc_3.5-2 > > loaded via a namespace (and not attached): > [1] cluster_1.11.12 grid_2.8.1 gtools_2.5.0-1 lattice_0.17-20 > [5] MASS_7.2-46 >> > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] change default output size when using Sweave
Dear Duncan, Thanks for the reply. This works, but unfortunately I need a different solution. My script is supposed to run completely automated and the graphics I produce vary in size each time I run the script. But I want the graphics to be fitted to my .pdf output without specifying the height argument manually each time. That is why I do not want a fixed height as a code chunk argument. Actually I do not know if it is possible to have a variable placed in a code chunk header. I tried the following which does not work: <<>>= size <- 3 @ <>= pushViewport(viewport(height = unit(80, "mm"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport() @ So still I face the problem to have Sweave generate a .pdf graphic that is just as big I want it to be. In the Sweave Docu paragraph A.9 I discovered something I use as a workaround. I produce the .pdf output manually (where I can control the size) and add each graphic to LaTex manually as well. <>= for (i in 1){ file=paste("myfile", i, ".pdf", sep="") pdf(file=file, paper="special", width=6, height=3) pushViewport(viewport(height = unit(5, "inches"))) grid.rect() grid.text("I want this viewport to be the whole output size") popViewport()dev.off() cat("\\includegraphics{", file, "}\n\n", sep="") } @ I use parenthesis around the code as it prints out something I do not want if no parenthesis are used (I use Windows). I am not too happy with the solution. I would prefer a more straightforward approach to define the size of the output graphic. I wonder if there are some Sweave settings that can be modified. In the Sweave manual (A.11) I found the following to customize the par settings for each figure: options(SweaveHooks=list(fig=function() par(bg="red", fg="blue"))) I wonder if something similar could be done changing the size of the default output device (pdf or eps) for each figure like >pdf.options(height=2) or similar (it seems that this does not work)? I suppose this type of graphic customization is quite a common issue when producing automated customized output/reports using R and Sweave but I haven't found anything concerning this topic yet. So I would be really glad if someone knows a solution. TIA, Mark -Ursprüngliche Nachricht- Von: Duncan Murdoch [mailto:murd...@stats.uwo.ca] Gesendet: Dienstag, 7. April 2009 19:08 An: Mark Heckmann Cc: r-help@r-project.org Betreff: Re: [R] change default output size when using Sweave On 4/7/2009 10:39 AM, Mark Heckmann wrote: > I am trying to create a graphic output in Sweave but I do not want it to be > standard size. I want the whole graphic to be 80mm of height only, just like > the viewport below. > > <>= >pushViewport(viewport(height = unit(80, "mm"))) >grid.rect() >grid.text("I want this viewport to be the whole output size") >popViewport() > @ > > How can I make the graphic smaller (80mm of height) without having the white > margins at the top and bottom? You can specify height and width as arguments to the code chunk or set new defaults with something like \SweaveOpts{width=5,height=3} (I think those measurements need to be given in inches; I don't think there's a way to use an expression like unit(80,"mm").) You can also control the size of the plot in the final document with LaTeX code like \usepackage{graphicx} \setkeys{Gin}{width=3in} and then the LaTeX rules for handling units would apply, so 80mm is fine instead of 3in. Duncan Murdoch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] does grid.rect() not accept NULL argument from gpar(col=NULL) ?
I want to draw a grid rectangle without a border. ?gpar says: "Specifying the value NULL for a parameter is the same as not specifying any value for that parameter, except for col and fill, where NULL indicates not to draw a border or not to fill an area (respectively)." pushViewport(viewport(height=unit(.8, "npc"))) grid.rect(gp=gpar(col=NULL, fill="green")) popViewport() Still a border is drawn. What am I doing wrong? TIA, Mark __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Null-Hypothesis
Hi Benedikt in principle this is correct. A p-value less than a prespecified alpha-level leads to rejection of the null hypothesis H0. I'm a little bit puzzled about your given p-value, since actually your slopes are the same, so H0 is valid but rejected. This might be a matter of chance, but it is so to say against all odds (~2 out of 10 000 000) having such a special data set. Alternatively - and using well established estimators - you can incorporate your factor in a joint model. df1 <- data.frame(x=1:3, y=1:3+rnorm(3),g=0) df2 <- data.frame(x=1:3, y=1:3+rnorm(3),g=1) dta<-rbind(df1,df2) dta$g<-factor(dta$g) # not really necessary for a two-group factor ll<-lm(y~x*g,dta) summary(ll) This model has a varying slope and intercept term (which is the same as fitting two separate regression lines as you did), the test for different slopes is to test significance of x:g. If you are in fact asking for a pure varying slope model you have to restrict your fit1 and fit2 to have the same intercept or you use the joint model with ll<-lm(y~x+x:g,dta) hth Benedikt Niesterok schrieb: Hello R users, I've used the following help two compare two regression line slopes. Wanted to test if they differ significantly: Hi, I've made a research about how to compare two regression line slopes (of y versus x for 2 groups, "group" being a factor ) using R. I knew the method based on the following statement : t = (b1 - b2) / sb1,b2 where b1 and b2 are the two slope coefficients and sb1,b2 the pooled standard error of the slope (b) which can be calculated in R this way: > df1 <- data.frame(x=1:3, y=1:3+rnorm(3)) > df2 <- data.frame(x=1:3, y=1:3+rnorm(3)) > fit1 <- lm(y~x, df1) > s1 <- summary(fit1)$coefficients > fit2 <- lm(y~x, df2) > s2 <- summary(fit2)$coefficients > db <- (s2[2,1]-s1[2,1]) > sd <- sqrt(s2[2,2]^2+s1[2,2]^2) > df <- (fit1$df.residual+fit2$df.residual) > td <- db/sd > 2*pt(-abs(td), df) My value I get by running this test is :[1] 2.305553e-07 Does it mean the two slopes differ significantly, because this value is in the alpha area, so that I have to reject the null- hypothesis and accept the alternative hypothesis? Is the null-hypothesis: slope1=slope2? Thanks for your help, Benedikt -- __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Eik Vettorazzi Institut für Medizinische Biometrie und Epidemiologie Universitätsklinikum Hamburg-Eppendorf Martinistr. 52 20246 Hamburg T ++49/40/42803-8243 F ++49/40/42803-7790 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Howto Plot With Transparent Background
with base graphics, par(bg=NA) see ?par Hope this helps, baptiste On 8 Apr 2009, at 12:54, Gundala Viswanath wrote: Is there a way to do it in R? Especially generating plot in EPS/PDF format. By transparent I mean clear (not white) background. I want to attached it to dark PPT slides. - Gundala Viswanath __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. _ Baptiste Auguié School of Physics University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Simulate binary data for a logistic regression Monte Carlo
Rolf Turner-3 wrote: > > > On 8/04/2009, at 1:27 PM, Ben Bolker wrote: > > > >> I agree that that the individual-level random effect is probably >> the issue. >> I played with this some today but didn't manage to resolve it -- >> tried JAGS/R2jags and glmer from lme4 but didn't manage to >> get an estimate of epsilon that matched the input value. I'm >> a little worried about binary data with an underlying random >> effect, I think there's an identifiability problem there ... > > Oh dear. This is getting hairier than I expected ... and a bit > over my head. The toy model proposed *looks* simple enough at > first blush. I would've thought that if a model which is that > simple doesn't ``work'', then what hope is there for the more > complicated models that one needs for analyzing real data? > > I don't think this is impossible, but it's a bit subtle, and might be worth taking up on r-sig-mixed-models instead ... Ben -- View this message in context: http://www.nabble.com/Simulate-binary-data-for-a-logistic-regression-Monte-Carlo-tp22928872p22949970.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Convert data frame containing time stamps to time series
Try varying the arguments to this to accommodate the precise format of your data. See the three zoo vignettes, ?read.zoo and R News 4/1 for dates and times. > Lines <- "t,f,p,d,o,s,a,l,c + 1/21/09 5:01,16151,8.2,76,30,282,1060,53,7 + 1/21/09 5:02,16256,8.3,76,23,282,1059,54,7 + 1/21/09 5:03,16150,8.4,76,26,282,1059,55,7 + 1/21/09 5:04,16150,9.0,76,25,282,1051,57,6 + 1/21/09 5:05,15543,10.4,76,7,282,1024,58,6" > library(zoo) > library(chron) > z <- read.zoo(textConnection(Lines), sep = ",", header = TRUE, + FUN = as.chron, format = "%m/%d/%y %H:%M") > z fp d o sa l c (01/21/09 05:01:00) 16151 8.2 76 30 282 1060 53 7 (01/21/09 05:02:00) 16256 8.3 76 23 282 1059 54 7 (01/21/09 05:03:00) 16150 8.4 76 26 282 1059 55 7 (01/21/09 05:04:00) 16150 9.0 76 25 282 1051 57 6 (01/21/09 05:05:00) 15543 10.4 76 7 282 1024 58 6 On Wed, Apr 8, 2009 at 4:43 AM, wrote: > I read records using scan: > > dat<-data.frame(scan(file="KDA.csv",what=list(t="%m/%d/%y > %H:%M",f=0,p=0,d=0,o=0,s=0,a=0,l=0,c=0),skip=2,sep=",",nmax=np,flush=TRUE,na.strings=c("I/OTimeout","ArcOff-line"))) > > which results in: > >> dat[1:5,] > t f p d o s a l c > 1 1/21/09 5:01 16151 8.2 76 30 282 1060 53 7 > 2 1/21/09 5:02 16256 8.3 76 23 282 1059 54 7 > 3 1/21/09 5:03 16150 8.4 76 26 282 1059 55 7 > 4 1/21/09 5:04 16150 9.0 76 25 282 1051 57 6 > 5 1/21/09 5:05 15543 10.4 76 7 282 1024 58 6 > > I have been unable to find a way to convert this into a time series. I did > read the manuals and came across a way to coerce a data frame to a ts > object: as.ts() > > Trouble is I do not know how to keep the timestamps in column t in the > data frame above. The t column is not strings. If I do: > > plot.ts(dat) > > I can see how the first graphics panel is indeed numbers not text. So I > think scan converted the text correctly per the format string I put in. > > Much more difficult still. The datafiles I have contain invalid data, > missing values and other none relevant information. I filter this out > using subset which works brilliantly. However, how can I filter using > subset and convert to a time series afterwards. Since after subsetting > there will be 'holes' i.e. missing records. Can a ts object deal with > missing records? If so, how? Just point me to a document. I can and will > put in the work to figure it out myself. > > Thank you! > Alex van der Spek > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Genstat into R - Randomisation test
Hello everybody, I have a question. I would like to get a correlation between constitutive and induced plant defence which I messured on 30 plant species. So I have table with Species, Induced defence (ID), and constitutive defence (CD). Since Induced and constitutive defence are not independant (so called spurious correlation) I should do a randomisation test. I have a syntax of my supervisor in Genstat, but I would really like to try this in R. "data from trade-off.IDCD" list variate [nval=1000] slope calc ID1=ID graph ID; CD calc b=corr(ID; CD) calc slope$[1]=b "slope$[1] is the correlation before permutating the data" for i=2...1000 randomize ID1 calc b=corr(CD1; ID1) calc slope$[i]=b endfor hist slope describe slope quantile [proportion=!(0.0005,0.005, 0.025, 0.975, 0.995,0.9995)]slope print slope$[1] corr[p=corr] ID,CD DHISTOGRAM [WINDOW=1; ORIENTATION=vertical; KEY=0; BARCOVERING=1.0] slope; PEN=2 Does anybody have done something similar and has any idea how to make the randomisation part? I would be very grateful for any help!! Thanks in advance, Anne -- Anne Kempel Institute of Plant Sciences University of Bern Altenbergrain 21 CH-3103 Bern kem...@ips.unibe.ch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.spss, locale and encodings
Hans Ekbrand wrote: I must be missing something obvious here: According to the help page for read.spss, the reencode option is only active when R is run under a UTF-8 locale. Not in my version: reencode: logical: should character strings be re-encoded to the current locale. The default, 'NA', means to do so in a UTF-8 locale, only. Alternatively character, specifying an encoding to assume. read.spss can only import the SPSS file when run under a iso88591(5) locale, under a UTF-8 locale I get: Error in read.spss("wo.sav") : error reading system-file header In addition: Warning message: In read.spss("wo.sav") : wo.sav: position 143: Variable name begins with invalid character So, does it help with reencode="Latin1"? Presumably this comes from assuming UTF-8 when it isn't. This is under Debian GNU/Linux, the stable release. foreign is version 8.27 8.34 is used in the current prerelease. AFAIR, some issues with encodings were fixed recently. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - (p.dalga...@biostat.ku.dk) FAX: (+45) 35327907 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Howto Plot With Transparent Background
if you not persist in using pdf, the emf format works very well with PPT and has a transparent background. Its the way you copy&paste pdf-graphs form Acrobat to PPT which causes the solid (white) background - unfortunately ppt has no builtt in support for pdf-graphs. By default pdfs have a transparent background, see ?pdf and there the bg part, which can be proven by a minimal R+ LaTeX example #R code pdf("test.pdf") plot(1,1) dev.off() #minimal tex example \documentclass[a4paper,12pt]{article} \usepackage[svgnames]{xcolor} \usepackage{graphicx} \begin{document} \pagecolor{green} \includegraphics[width=.5\textwidth]{test.pdf} \end{document} hth. Gundala Viswanath schrieb: Is there a way to do it in R? Especially generating plot in EPS/PDF format. By transparent I mean clear (not white) background. I want to attached it to dark PPT slides. - Gundala Viswanath __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Eik Vettorazzi Institut für Medizinische Biometrie und Epidemiologie Universitätsklinikum Hamburg-Eppendorf Martinistr. 52 20246 Hamburg T ++49/40/42803-8243 F ++49/40/42803-7790 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] factor, as.factor and levels
It is just a simple version of 'factor'. The only speed advantage it might have is that it checks to see if it is a factor first. Here is the definition: )> as.factor function (x) if (is.factor(x)) x else factor(x) You can always list out what the function does to get a better understanding of how it works. On Wed, Apr 8, 2009 at 8:16 AM, Heinz Tuechler wrote: > Thank you, Jim. I see, the fact that in the documentation you find only > "as.factor(x)" means that it does not accept more arguments. > Does as.factor have speed advantages over factor, or is there a different > cause for it's existence? > > Heinz > > > > At 13:50 08.04.2009, jim holtman wrote: >> >> as.factor does not accept levels as an argument. use the first form >> that you have >> >> factor(ch1, levels=ch1) >> >> >> >> On Wed, Apr 8, 2009 at 7:36 AM, Heinz Tuechler wrote: >> > Dear All, >> > >> > to my surprise as.factor does not accept a levels argument. Maybe I did >> > not >> > read the documentation well enough. See the example below. I wanted to >> > use >> > ch1 as factor in the newdata argument of survfit, so I assumed that I >> > could >> > write as.factor(ch1, levels=ch1), since the order should be kept. >> > >> > But as.factor(ch1, levels=ch1) results in the error: >> > >> > Error in as.factor(ch1, levels = ch1) : >> > unused argument(s) (levels = c("low", "inter", "high")) >> > >> > factor(ch1, levels=ch1) works as I expected. >> > Is it intended that as.factor does not use the levels argument? >> > >> > Thanks, >> > >> > Heinz >> > >> > ch1 <- c('low', 'inter', 'high') >> > factor(ch1) >> > factor(ch1, levels=ch1) >> > as.factor(ch1, levels=ch1) >> > >> >> version >> > _ >> > platform i386-pc-mingw32 >> > arch i386 >> > os mingw32 >> > system i386, mingw32 >> > status Patched >> > major 2 >> > minor 8.1 >> > year 2009 >> > month 03 >> > day 13 >> > svn rev 48132 >> > language R >> > version.string R version 2.8.1 Patched (2009-03-13 r48132) >> >> sessionInfo() >> > R version 2.8.1 Patched (2009-03-13 r48132) >> > i386-pc-mingw32 >> > >> > locale: >> > >> > LC_COLLATE=German_Switzerland.1252;LC_CTYPE=German_Switzerland.1252;LC_MONETARY=German_Switzerland.1252;LC_NUMERIC=C;LC_TIME=German_Switzerland.1252 >> > >> > attached base packages: >> > [1] splines stats graphics grDevices utils datasets methods >> > [8] base >> > >> > other attached packages: >> > [1] survival_2.34-1 car_1.2-12 gmodels_2.14.1 gdata_2.4.2 >> > [5] Hmisc_3.5-2 >> > >> > loaded via a namespace (and not attached): >> > [1] cluster_1.11.12 grid_2.8.1 gtools_2.5.0-1 lattice_0.17-20 >> > [5] MASS_7.2-46 >> >> >> > >> > __ >> > R-help@r-project.org mailing list >> > https://stat.ethz.ch/mailman/listinfo/r-help >> > PLEASE do read the posting guide >> > http://www.R-project.org/posting-guide.html >> > and provide commented, minimal, self-contained, reproducible code. >> > >> >> >> >> -- >> Jim Holtman >> Cincinnati, OH >> +1 513 646 9390 >> >> What is the problem that you are trying to solve? > > > -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.spss, locale and encodings
On Wed, Apr 08, 2009 at 03:03:06PM +0200, Peter Dalgaard wrote: > Hans Ekbrand wrote: >> I must be missing something obvious here: >> >> According to the help page for read.spss, the reencode option is only >> active when R is run under a UTF-8 locale. > > Not in my version: > > reencode: logical: should character strings be re-encoded to the > current locale. The default, 'NA', means to do so in a UTF-8 > locale, only. Alternatively character, specifying an > encoding to assume. OK, thanks for that correction, but the problem isn't solved, since read.spss fails, see below. When read.spss succeeds, the options is not useful, since then the current locale is iso88591(5). > So, does it help with reencode="Latin1"? Presumably this comes from > assuming UTF-8 when it isn't. > Sys.getlocale() [1] "LC_CTYPE=sv_SE.UTF-8;LC_NUMERIC=C;LC_TIME=sv_SE.UTF-8;LC_COLLATE=sv_SE.UTF-8;LC_MONETARY=sv_SE.UTF-8;LC_MESSAGES=sv_SE.utf8;LC_PAPER=sv_SE.utf8;LC_NAME=C;LC_ADDRESS=C;LC_TELEPHONE=C;LC_MEASUREMENT=sv_SE.utf8;LC_IDENTIFICATION=C" > test <- read.spss("wo.sav", to.data.frame=TRUE, reencode="Latin1") Error in read.spss("wo.sav", to.data.frame = TRUE, reencode = "Latin1") : error reading system-file header In addition: Warning message: In read.spss("wo.sav", to.data.frame = TRUE, reencode = "Latin1") : wo.sav: position 143: Variable name begins with invalid character Using another version of the dataset, where I have successfully encoded the names to UTF-8, here is the problematic variable name: names(Workorientation.2005.Swe)[143] [1] "KÖN1" > 8.34 is used in the current prerelease. AFAIR, some issues with > encodings were fixed recently. Someone running foreign 8.34 that is willing to test my SPSS-file? -- Hans Ekbrand (http://sociologi.cjb.net) Q. What is that strange attachment in this mail? A. My digital signature, see www.gnupg.org for info on how you could use it to ensure that this mail is from me and has not been altered on the way to you. signature.asc Description: Digital signature __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] factor, as.factor and levels
Jim - you are right, I should have looked before. So there is a difference that should also effect the dropping of unused levels. Thanks, Heinz At 15:31 08.04.2009, jim holtman wrote: It is just a simple version of 'factor'. The only speed advantage it might have is that it checks to see if it is a factor first. Here is the definition: )> as.factor function (x) if (is.factor(x)) x else factor(x) You can always list out what the function does to get a better understanding of how it works. On Wed, Apr 8, 2009 at 8:16 AM, Heinz Tuechler wrote: > Thank you, Jim. I see, the fact that in the documentation you find only > "as.factor(x)" means that it does not accept more arguments. > Does as.factor have speed advantages over factor, or is there a different > cause for it's existence? > > Heinz > > > > At 13:50 08.04.2009, jim holtman wrote: >> >> as.factor does not accept levels as an argument. use the first form >> that you have >> >> factor(ch1, levels=ch1) >> >> >> >> On Wed, Apr 8, 2009 at 7:36 AM, Heinz Tuechler wrote: >> > Dear All, >> > >> > to my surprise as.factor does not accept a levels argument. Maybe I did >> > not >> > read the documentation well enough. See the example below. I wanted to >> > use >> > ch1 as factor in the newdata argument of survfit, so I assumed that I >> > could >> > write as.factor(ch1, levels=ch1), since the order should be kept. >> > >> > But as.factor(ch1, levels=ch1) results in the error: >> > >> > Error in as.factor(ch1, levels = ch1) : >> > unused argument(s) (levels = c("low", "inter", "high")) >> > >> > factor(ch1, levels=ch1) works as I expected. >> > Is it intended that as.factor does not use the levels argument? >> > >> > Thanks, >> > >> > Heinz >> > >> > ch1 <- c('low', 'inter', 'high') >> > factor(ch1) >> > factor(ch1, levels=ch1) >> > as.factor(ch1, levels=ch1) >> > >> >> version >> > _ >> > platform i386-pc-mingw32 >> > arch i386 >> > os mingw32 >> > system i386, mingw32 >> > status Patched >> > major 2 >> > minor 8.1 >> > year 2009 >> > month 03 >> > day13 >> > svn rev48132 >> > language R >> > version.string R version 2.8.1 Patched (2009-03-13 r48132) >> >> sessionInfo() >> > R version 2.8.1 Patched (2009-03-13 r48132) >> > i386-pc-mingw32 >> > >> > locale: >> > >> > LC_COLLATE=German_Switzerland.1252;LC_CTYPE=German_Switzerland.1252;LC_MONETARY=German_Switzerland.1252;LC_NUMERIC=C;LC_TIME=German_Switzerland.1252 >> > >> > attached base packages: >> > [1] splines stats graphics grDevices utils datasets methods >> > [8] base >> > >> > other attached packages: >> > [1] survival_2.34-1 car_1.2-12 gmodels_2.14.1 gdata_2.4.2 >> > [5] Hmisc_3.5-2 >> > >> > loaded via a namespace (and not attached): >> > [1] cluster_1.11.12 grid_2.8.1 gtools_2.5.0-1 lattice_0.17-20 >> > [5] MASS_7.2-46 >> >> >> > >> > __ >> > R-help@r-project.org mailing list >> > https://stat.ethz.ch/mailman/listinfo/r-help >> > PLEASE do read the posting guide >> > http://www.R-project.org/posting-guide.html >> > and provide commented, minimal, self-contained, reproducible code. >> > >> >> >> >> -- >> Jim Holtman >> Cincinnati, OH >> +1 513 646 9390 >> >> What is the problem that you are trying to solve? > > > -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] sourced plot commands not working
I have source a script running with no errors that ends in these lines: == for (i in 2:3) win.graph() dev.set(2) levelplot(avg ~ procs * size, rrt, drape=T, colorkey=T,main="ops/sec",scales=list(x=list(tick.number=15),y=list(log=10))) dev.set(3) levelplot(avg ~ procs * size, ort, drape=T, colorkey=T,main="ops/sec",scales=list(x=list(tick.number=15),y=list(log=10))) = The graphics devices windows appear, however, there is no graph and the background of each window is gray indicating no active graph. However, if I copy and paste these commands from the script and enter them into the console, they run fine. why? R 2.8.1 2008-12-22 on windows. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Pulling data into a model
Hi Jacy, I'll take my chances answering your question. I'd suggest using the paste() function and then transforming the string into a formula object: my.formula = as.formula(paste("y~x1+x2+",paste(colnames(REG)[2:num],collapse="+"), sep= "")) ; aov(my.formula,data=REG) ; The first line creates a string that then becomes your formula statement. Is this what you're looking for? Cheers, Luc Crosby, Jacy R wrote: I'm creating an ANOVA model. Is there a way to pull in consecutive columns of variables for the test? Example: aov1<-aov(y~x1+x2+REG[,2:num], data=REG) I'm not looking for interaction effects, I just want to create a model for the first few columns of variables (exact number and names will vary) and a few other predetermined variables (in the example I named them x1 and x2). Thanks in advance. - Jacy [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] R File I/O Capability - Writing output to specific lines of existing file
Currently I am using the R "write" command to output results to a *.txt file and then copying those results into an initialization file. In an attempt to continue to automate the process I would like to have R write to the location in the existing initialization file, instead of me copying the data over. By any chance are there any R commands to help? Primarily, I will be using Windows, and the initialization file is a simple flat file, i.e. not XML or binary. I looked at: (a) http://www.stat.ucl.ac.be/ISdidactique/Rhelp/library/R.io/html/00Index.html (b) http://search.r-project.org/cgi-bin/namazu.cgi?query=File+I%2FO&max=100&result=normal&sort=score&idxname=functions&idxname=Rhelp08 But, this did not appear to provide the functionality to edit an existing file by adding information to the middle of the file. Thank you again for any help and insight. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] sourced plot commands not working
On Wed, 2009-04-08 at 10:05 -0400, SHANE MILLER, BLOOMBERG/ 731 LEXIN wrote: > I have source a script running with no errors that ends in these lines: > == > for (i in 2:3) win.graph() > dev.set(2) > levelplot(avg ~ procs * size, rrt, drape=T, > colorkey=T,main="ops/sec",scales=list(x=list(tick.number=15),y=list(log=10))) > dev.set(3) > levelplot(avg ~ procs * size, ort, drape=T, > colorkey=T,main="ops/sec",scales=list(x=list(tick.number=15),y=list(log=10))) > = > The graphics devices windows appear, however, there is no graph > and the background of each window is gray indicating no active > graph. > > However, if I copy and paste these commands from the script and > enter them into the console, they run fine. why? The plotting commands are lattice functions. These need to be printed to produce the plotting side effect you want. When entered at the command line printing is explicit. This doesn't happen when sourced. This is a FAQ 7.22: > > R 2.8.1 2008-12-22 on windows. > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Dr. Gavin Simpson [t] +44 (0)20 7679 0522 ECRC, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/ UK. WC1E 6BT. [w] http://www.freshwaters.org.uk %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] default print format for large numbers
Numbers like ``1239178547.653775" is inserted into a vector. I print the vector: > route_9_80_end [1] 1239178522 1239178526 1239178524 1239178524 1239178524 1239178523 1239178524 1239178522 1239178521 1239178565 1239178566 1239178566 [13] 1239178565 1239178566 1239178566 1239178565 1239178566 1239178566 and the decimals are suppressed. While R is wonderful and wonderfully documented, I can only find some quasi relavent information on options(digits) or some such as might apply here. How can I force R to print some part of the decimal i.e. back to 1239178547.653775? TIA, Shane __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.spss, locale and encodings
Hans Ekbrand wrote: On Wed, Apr 08, 2009 at 03:03:06PM +0200, Peter Dalgaard wrote: Hans Ekbrand wrote: I must be missing something obvious here: According to the help page for read.spss, the reencode option is only active when R is run under a UTF-8 locale. Not in my version: reencode: logical: should character strings be re-encoded to the current locale. The default, 'NA', means to do so in a UTF-8 locale, only. Alternatively character, specifying an encoding to assume. OK, thanks for that correction, but the problem isn't solved, since read.spss fails, see below. When read.spss succeeds, the options is not useful, since then the current locale is iso88591(5). So, does it help with reencode="Latin1"? Presumably this comes from assuming UTF-8 when it isn't. Sys.getlocale() [1] "LC_CTYPE=sv_SE.UTF-8;LC_NUMERIC=C;LC_TIME=sv_SE.UTF-8;LC_COLLATE=sv_SE.UTF-8;LC_MONETARY=sv_SE.UTF-8;LC_MESSAGES=sv_SE.utf8;LC_PAPER=sv_SE.utf8;LC_NAME=C;LC_ADDRESS=C;LC_TELEPHONE=C;LC_MEASUREMENT=sv_SE.utf8;LC_IDENTIFICATION=C" test <- read.spss("wo.sav", to.data.frame=TRUE, reencode="Latin1") Error in read.spss("wo.sav", to.data.frame = TRUE, reencode = "Latin1") : error reading system-file header In addition: Warning message: In read.spss("wo.sav", to.data.frame = TRUE, reencode = "Latin1") : wo.sav: position 143: Variable name begins with invalid character Using another version of the dataset, where I have successfully encoded the names to UTF-8, here is the problematic variable name: names(Workorientation.2005.Swe)[143] [1] "KÖN1" 8.34 is used in the current prerelease. AFAIR, some issues with encodings were fixed recently. Someone running foreign 8.34 that is willing to test my SPSS-file? Someone with an SPSS file problem willing to help test the prereleases? :-) You could start by placing it somewhere accessible... -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - (p.dalga...@biostat.ku.dk) FAX: (+45) 35327907 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] sourced plot commands not working
[Ooops, brain/fingers not well connected today, sent prematurely] On Wed, 2009-04-08 at 10:05 -0400, SHANE MILLER, BLOOMBERG/ 731 LEXIN wrote: > I have source a script running with no errors that ends in these lines: > == > for (i in 2:3) win.graph() > dev.set(2) > levelplot(avg ~ procs * size, rrt, drape=T, > colorkey=T,main="ops/sec",scales=list(x=list(tick.number=15),y=list(log=10))) > dev.set(3) > levelplot(avg ~ procs * size, ort, drape=T, > colorkey=T,main="ops/sec",scales=list(x=list(tick.number=15),y=list(log=10))) > = > The graphics devices windows appear, however, there is no graph > and the background of each window is gray indicating no active > graph. > > However, if I copy and paste these commands from the script and > enter them into the console, they run fine. why? The plotting commands used here are lattice functions. These need to be printed to produce the plotting side effect you want. When entered at the command line printing is explicit. This doesn't happen when sourced. This is a FAQ 7.22: http://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-do-lattice_002ftrellis-graphics-not-work_003f ?source does explain all of this. Solution; wrap your levelplot(...) in print(): print(levelplot(...)) HTH G > > R 2.8.1 2008-12-22 on windows. > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Dr. Gavin Simpson [t] +44 (0)20 7679 0522 ECRC, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/ UK. WC1E 6BT. [w] http://www.freshwaters.org.uk %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] binary version of R 2.8.x
./configure works with no errors ? I have compiled on mac os x and debian so my advice may be circumspect, but the errors I don't know what they are. On Tue, Apr 7, 2009 at 10:17 PM, wrote: > > > Hello, > > Is there any binary version of R 2.8.x available for x86 Solaris 10? I need > it so that I can use some R library files. I actually tried to compile the > source myself, but it failed. I don't know much about how to compile source > files on Solaris, and it will be great if there is a binary version of R > 2.8.x available. Thanks. > > /***error after running make***/ > util.c: In function `Rf_mbrtowc': > util.c:1147: parse error before `char' > util.c:1150: `p' undeclared (first use in this function) > util.c:1150: (Each undeclared identifier is reported only once > util.c:1150: for each function it appears in.) > util.c:1150: `q' undeclared (first use in this function) > util.c:1150: `err' undeclared (first use in this function) > make[3]: *** [util.o] Error 1 > make[3]: Leaving directory `/var/tmp/R-2.8.1/src/main' > make[2]: *** [R] Error 2 > make[2]: Leaving directory `/var/tmp/R-2.8.1/src/main' > make[1]: *** [R] Error 1 > make[1]: Leaving directory `/var/tmp/R-2.8.1/src' > make: *** [R] Error 1 > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Stephen Sefick Let's not spend our time and resources thinking about things that are so little or so large that all they really do for us is puff us up and make us feel like gods. We are mammals, and have not exhausted the annoying little problems of being mammals. -K. Mullis __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Re ading Excel 5.0 files with RODBC?
Seems like the place where i get those files from has an error somewhere when they create the file which doesn't allow ODBC to see sheet with data, i.e sqlTables returns: [1] TABLE_CAT TABLE_SCHEM TABLE_NAME TABLE_TYPE REMARKS <0 rows> (or 0-length row.names) When i open this file in excel 2003 and try to save it, i get a message that this file is of Excel 5.0 format (but the original file has more than 16k rows which is impossible with excel 5.0 format) and information wil be lost. When i save it on my pc either in excel 5.0 or excel 97 format i can open it with RODBC with no problems so the problem has to be with the way original xls files are created. Thanks a lot for all replies. Prof Brian Ripley wrote: > > On Wed, 8 Apr 2009, Yuri Volchik wrote: > >> >> Thanks for reply David, >> >> but the problem is that the place i get those files from saves them in >> the >> old Excel format (Excel 5.0), not Excel 97-2003, so neither of the >> packages >> (RODBC, xlsReadWrite) can read it. > > RODBC supports any format for which an ODBC driver is installed and > Microsoft have long supplied ones for Excel 3.0 and later (up to Excel > 2007 at present). > > -- > Brian D. Ripley, rip...@stats.ox.ac.uk > Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ > University of Oxford, Tel: +44 1865 272861 (self) > 1 South Parks Road, +44 1865 272866 (PA) > Oxford OX1 3TG, UKFax: +44 1865 272595 > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > > -- View this message in context: http://www.nabble.com/Reading-Excel-5.0-files-with-RODBC--tp22929781p22951583.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Genstat into R - Randomisation test
Looks like that code implements a non-exhaustive variant of the randomization test, sometimes called a permutation test. Assuming you have a data frame with columns ID & CD, this should do it: n.obs = 100 ID=rnorm(n.obs) CD=rnorm(n.obs) obs.cor = cor(ID,CD) num.permutations = 1e4 perm.cor = rep(NA,num.permutations) start.time=proc.time()[1] index = 1:n.obs for(i in 1:num.permutations){ IDorder = sample(index) CDorder = sample(index) perm.cor[i] = .Internal(cor(ID[IDorder], CD[CDorder], 4, FALSE)) } cat('Elapsed time:',start.time-proc.time(1)) sum(perm.cor>obs.cor)/num.permutations On Wed, Apr 8, 2009 at 10:02 AM, Anne Kempel wrote: > Hello everybody, > I have a question. I would like to get a correlation between constitutive > and induced plant defence which I messured on 30 plant species. So I have > table with Species, Induced defence (ID), and constitutive defence (CD). > Since Induced and constitutive defence are not independant (so called > spurious correlation) I should do a randomisation test. I have a syntax of > my supervisor in Genstat, but I would really like to try this in R. > > > "data from trade-off.IDCD" > list > variate [nval=1000] slope > calc ID1=ID > > graph ID; CD > calc b=corr(ID; CD) > calc slope$[1]=b > > "slope$[1] is the correlation before permutating the data" > > for i=2...1000 > randomize ID1 > calc b=corr(CD1; ID1) > calc slope$[i]=b > endfor > > hist slope > describe slope > quantile [proportion=!(0.0005,0.005, 0.025, 0.975, 0.995,0.9995)]slope > print slope$[1] > corr[p=corr] ID,CD > > > DHISTOGRAM [WINDOW=1; ORIENTATION=vertical; KEY=0; BARCOVERING=1.0] slope; > PEN=2 > > Does anybody have done something similar and has any idea how to make the > randomisation part? > I would be very grateful for any help!! > Thanks in advance, > Anne > > > > > -- > Anne Kempel > Institute of Plant Sciences > University of Bern > Altenbergrain 21 > CH-3103 Bern > kem...@ips.unibe.ch > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University Looking to arrange a meeting? Check my public calendar: http://tinyurl.com/mikes-public-calendar ~ Certainty is folly... I think. ~ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Genstat into R - Randomisation test
"Assuming you have a data frame with columns ID & CD, this should do it" Oops, the code I sent doesn't assume this. It assumes that you have two vectors, ID & CD, as generated in the first 3 lines. On Wed, Apr 8, 2009 at 11:31 AM, Mike Lawrence wrote: > Looks like that code implements a non-exhaustive variant of the > randomization test, sometimes called a permutation test. Assuming you > have a data frame with columns ID & CD, this should do it: > > n.obs = 100 > ID=rnorm(n.obs) > CD=rnorm(n.obs) > > obs.cor = cor(ID,CD) > > num.permutations = 1e4 > perm.cor = rep(NA,num.permutations) > start.time=proc.time()[1] > index = 1:n.obs > for(i in 1:num.permutations){ > IDorder = sample(index) > CDorder = sample(index) > perm.cor[i] = .Internal(cor(ID[IDorder], CD[CDorder], 4, FALSE)) > } > cat('Elapsed time:',start.time-proc.time(1)) > sum(perm.cor>obs.cor)/num.permutations > > > On Wed, Apr 8, 2009 at 10:02 AM, Anne Kempel wrote: >> Hello everybody, >> I have a question. I would like to get a correlation between constitutive >> and induced plant defence which I messured on 30 plant species. So I have >> table with Species, Induced defence (ID), and constitutive defence (CD). >> Since Induced and constitutive defence are not independant (so called >> spurious correlation) I should do a randomisation test. I have a syntax of >> my supervisor in Genstat, but I would really like to try this in R. >> >> >> "data from trade-off.IDCD" >> list >> variate [nval=1000] slope >> calc ID1=ID >> >> graph ID; CD >> calc b=corr(ID; CD) >> calc slope$[1]=b >> >> "slope$[1] is the correlation before permutating the data" >> >> for i=2...1000 >> randomize ID1 >> calc b=corr(CD1; ID1) >> calc slope$[i]=b >> endfor >> >> hist slope >> describe slope >> quantile [proportion=!(0.0005,0.005, 0.025, 0.975, 0.995,0.9995)]slope >> print slope$[1] >> corr[p=corr] ID,CD >> >> >> DHISTOGRAM [WINDOW=1; ORIENTATION=vertical; KEY=0; BARCOVERING=1.0] slope; >> PEN=2 >> >> Does anybody have done something similar and has any idea how to make the >> randomisation part? >> I would be very grateful for any help!! >> Thanks in advance, >> Anne >> >> >> >> >> -- >> Anne Kempel >> Institute of Plant Sciences >> University of Bern >> Altenbergrain 21 >> CH-3103 Bern >> kem...@ips.unibe.ch >> >> __ >> R-help@r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >> > > > > -- > Mike Lawrence > Graduate Student > Department of Psychology > Dalhousie University > > Looking to arrange a meeting? Check my public calendar: > http://tinyurl.com/mikes-public-calendar > > ~ Certainty is folly... I think. ~ > -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University Looking to arrange a meeting? Check my public calendar: http://tinyurl.com/mikes-public-calendar ~ Certainty is folly... I think. ~ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Genstat into R - Randomisation test
*Hits head* Of course, the approach taken by your Genstat code of only shuffling one variable is sufficient and faster: n.obs = 100 ID=rnorm(n.obs) CD=rnorm(n.obs) obs.cor = cor(ID,CD) num.permutations = 1e4 perm.cor = rep(NA,num.permutations) start.time=proc.time()[1] index = 1:n.obs for(i in 1:num.permutations){ IDorder = sample(index) perm.cor[i] = .Internal(cor(ID[IDorder], CD, 4, FALSE)) } cat('Elapsed time:',start.time-proc.time(1)) sum(perm.cor>obs.cor)/num.permutations On Wed, Apr 8, 2009 at 11:33 AM, Mike Lawrence wrote: > "Assuming you have a data frame with columns ID & CD, this should do it" > > Oops, the code I sent doesn't assume this. It assumes that you have > two vectors, ID & CD, as generated in the first 3 lines. > > On Wed, Apr 8, 2009 at 11:31 AM, Mike Lawrence wrote: >> Looks like that code implements a non-exhaustive variant of the >> randomization test, sometimes called a permutation test. Assuming you >> have a data frame with columns ID & CD, this should do it: >> >> n.obs = 100 >> ID=rnorm(n.obs) >> CD=rnorm(n.obs) >> >> obs.cor = cor(ID,CD) >> >> num.permutations = 1e4 >> perm.cor = rep(NA,num.permutations) >> start.time=proc.time()[1] >> index = 1:n.obs >> for(i in 1:num.permutations){ >> IDorder = sample(index) >> CDorder = sample(index) >> perm.cor[i] = .Internal(cor(ID[IDorder], CD[CDorder], 4, FALSE)) >> } >> cat('Elapsed time:',start.time-proc.time(1)) >> sum(perm.cor>obs.cor)/num.permutations >> >> >> On Wed, Apr 8, 2009 at 10:02 AM, Anne Kempel wrote: >>> Hello everybody, >>> I have a question. I would like to get a correlation between constitutive >>> and induced plant defence which I messured on 30 plant species. So I have >>> table with Species, Induced defence (ID), and constitutive defence (CD). >>> Since Induced and constitutive defence are not independant (so called >>> spurious correlation) I should do a randomisation test. I have a syntax of >>> my supervisor in Genstat, but I would really like to try this in R. >>> >>> >>> "data from trade-off.IDCD" >>> list >>> variate [nval=1000] slope >>> calc ID1=ID >>> >>> graph ID; CD >>> calc b=corr(ID; CD) >>> calc slope$[1]=b >>> >>> "slope$[1] is the correlation before permutating the data" >>> >>> for i=2...1000 >>> randomize ID1 >>> calc b=corr(CD1; ID1) >>> calc slope$[i]=b >>> endfor >>> >>> hist slope >>> describe slope >>> quantile [proportion=!(0.0005,0.005, 0.025, 0.975, 0.995,0.9995)]slope >>> print slope$[1] >>> corr[p=corr] ID,CD >>> >>> >>> DHISTOGRAM [WINDOW=1; ORIENTATION=vertical; KEY=0; BARCOVERING=1.0] slope; >>> PEN=2 >>> >>> Does anybody have done something similar and has any idea how to make the >>> randomisation part? >>> I would be very grateful for any help!! >>> Thanks in advance, >>> Anne >>> >>> >>> >>> >>> -- >>> Anne Kempel >>> Institute of Plant Sciences >>> University of Bern >>> Altenbergrain 21 >>> CH-3103 Bern >>> kem...@ips.unibe.ch >>> >>> __ >>> R-help@r-project.org mailing list >>> https://stat.ethz.ch/mailman/listinfo/r-help >>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html >>> and provide commented, minimal, self-contained, reproducible code. >>> >> >> >> >> -- >> Mike Lawrence >> Graduate Student >> Department of Psychology >> Dalhousie University >> >> Looking to arrange a meeting? Check my public calendar: >> http://tinyurl.com/mikes-public-calendar >> >> ~ Certainty is folly... I think. ~ >> > > > > -- > Mike Lawrence > Graduate Student > Department of Psychology > Dalhousie University > > Looking to arrange a meeting? Check my public calendar: > http://tinyurl.com/mikes-public-calendar > > ~ Certainty is folly... I think. ~ > -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University Looking to arrange a meeting? Check my public calendar: http://tinyurl.com/mikes-public-calendar ~ Certainty is folly... I think. ~ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] default print format for large numbers
> Numbers like ``1239178547.653775" is inserted into a vector. I print > the vector: > > > route_9_80_end > [1] 1239178522 1239178526 1239178524 1239178524 1239178524 > 1239178523 1239178524 1239178522 1239178521 1239178565 1239178566 1239178566 > [13] 1239178565 1239178566 1239178566 1239178565 1239178566 1239178566 > > and the decimals are suppressed. While R is wonderful and > wonderfully documented, I can only find some quasi relavent > information on options(digits) or some such as might apply here. How > can I force R to print some part of the decimal i.e. back to > 1239178547.653775? When you create the vector, each number is stored as a double-precision floating point number. When you type the name of the variable the print function is called, which displays the numbers on the console. You can also manually call the print function, and set the digits argument to a high number to show all digits. If you want to see all the digits every time any number is displayed, use options(digits). For example: x <- 1239178547.653775 x # [1] 1239178548 print(x) # [1] 1239178548 print(x, digits=20) # [1] 1239178547.653775 options(digits=20) x # [1] 1239178547.653775 Note that the accuracy of the floating point numbers is limited to .Machine$double.eps, which is typically 2.2e-16. Regards, Richie. Mathematical Sciences Unit HSL ATTENTION: This message contains privileged and confidential inform...{{dropped:20}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Genstat into R - Randomisation test
*sigh* and the elapsed time line should be: cat('Elapsed time:',proc.time()[1]-start.time) On Wed, Apr 8, 2009 at 11:35 AM, Mike Lawrence wrote: > *Hits head* > Of course, the approach taken by your Genstat code of only shuffling > one variable is sufficient and faster: > > n.obs = 100 > ID=rnorm(n.obs) > CD=rnorm(n.obs) > > obs.cor = cor(ID,CD) > > num.permutations = 1e4 > perm.cor = rep(NA,num.permutations) > start.time=proc.time()[1] > index = 1:n.obs > for(i in 1:num.permutations){ > IDorder = sample(index) > perm.cor[i] = .Internal(cor(ID[IDorder], CD, 4, FALSE)) > } > cat('Elapsed time:',start.time-proc.time(1)) > sum(perm.cor>obs.cor)/num.permutations > > > On Wed, Apr 8, 2009 at 11:33 AM, Mike Lawrence wrote: >> "Assuming you have a data frame with columns ID & CD, this should do it" >> >> Oops, the code I sent doesn't assume this. It assumes that you have >> two vectors, ID & CD, as generated in the first 3 lines. >> >> On Wed, Apr 8, 2009 at 11:31 AM, Mike Lawrence wrote: >>> Looks like that code implements a non-exhaustive variant of the >>> randomization test, sometimes called a permutation test. Assuming you >>> have a data frame with columns ID & CD, this should do it: >>> >>> n.obs = 100 >>> ID=rnorm(n.obs) >>> CD=rnorm(n.obs) >>> >>> obs.cor = cor(ID,CD) >>> >>> num.permutations = 1e4 >>> perm.cor = rep(NA,num.permutations) >>> start.time=proc.time()[1] >>> index = 1:n.obs >>> for(i in 1:num.permutations){ >>> IDorder = sample(index) >>> CDorder = sample(index) >>> perm.cor[i] = .Internal(cor(ID[IDorder], CD[CDorder], 4, FALSE)) >>> } >>> cat('Elapsed time:',start.time-proc.time(1)) >>> sum(perm.cor>obs.cor)/num.permutations >>> >>> >>> On Wed, Apr 8, 2009 at 10:02 AM, Anne Kempel wrote: Hello everybody, I have a question. I would like to get a correlation between constitutive and induced plant defence which I messured on 30 plant species. So I have table with Species, Induced defence (ID), and constitutive defence (CD). Since Induced and constitutive defence are not independant (so called spurious correlation) I should do a randomisation test. I have a syntax of my supervisor in Genstat, but I would really like to try this in R. "data from trade-off.IDCD" list variate [nval=1000] slope calc ID1=ID graph ID; CD calc b=corr(ID; CD) calc slope$[1]=b "slope$[1] is the correlation before permutating the data" for i=2...1000 randomize ID1 calc b=corr(CD1; ID1) calc slope$[i]=b endfor hist slope describe slope quantile [proportion=!(0.0005,0.005, 0.025, 0.975, 0.995,0.9995)]slope print slope$[1] corr[p=corr] ID,CD DHISTOGRAM [WINDOW=1; ORIENTATION=vertical; KEY=0; BARCOVERING=1.0] slope; PEN=2 Does anybody have done something similar and has any idea how to make the randomisation part? I would be very grateful for any help!! Thanks in advance, Anne -- Anne Kempel Institute of Plant Sciences University of Bern Altenbergrain 21 CH-3103 Bern kem...@ips.unibe.ch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. >>> >>> >>> >>> -- >>> Mike Lawrence >>> Graduate Student >>> Department of Psychology >>> Dalhousie University >>> >>> Looking to arrange a meeting? Check my public calendar: >>> http://tinyurl.com/mikes-public-calendar >>> >>> ~ Certainty is folly... I think. ~ >>> >> >> >> >> -- >> Mike Lawrence >> Graduate Student >> Department of Psychology >> Dalhousie University >> >> Looking to arrange a meeting? Check my public calendar: >> http://tinyurl.com/mikes-public-calendar >> >> ~ Certainty is folly... I think. ~ >> > > > > -- > Mike Lawrence > Graduate Student > Department of Psychology > Dalhousie University > > Looking to arrange a meeting? Check my public calendar: > http://tinyurl.com/mikes-public-calendar > > ~ Certainty is folly... I think. ~ > -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University Looking to arrange a meeting? Check my public calendar: http://tinyurl.com/mikes-public-calendar ~ Certainty is folly... I think. ~ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] persp3d and rgl.viewpoint for rotating 3D plots
Dear R-users, within the rgl-package, I would have a question about the usage of persp3d in combination of rgl.viewpoint. I am not able to figure out how to let a 3D plot rotating around likewise the example in ?rgl.viewpoint. It seems that when I use persp3d(...) I see something on my screen, which is different from what I get when it's rotating. Is there any different behavior between shade3d and persp3d? Please find below a simple example. Actually, my final aim is to save each of the rotating graphs for creating a "clip". I commented below what I would use later on, but is there any way to do it better? Thanks a lot for your help, Carlo Giovanni Camarda library(rgl) # simplified version from help(rgl.viewpoint) shade3d(oh3d()) coo <- 1:360 for(i in 1:length(coo)) { rgl.viewpoint(coo[i]) #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") #rgl.snapshot(filename) } # simple persp3d adaption f <- function(x,y) { r <- sqrt(x^2+y^2); 10 * sin(r)/r } x <- seq(-10, 10, length= 30); y <- x z <- outer(x, y, f) persp3d(x, y, z) coo <- 1:360 for(i in 1:length(coo)) { rgl.viewpoint(coo[i]) #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") #rgl.snapshot(filename) } - Dr. Carlo Giovanni Camarda Research Scientist Max Planck Institute for Demographic Research Laboratory of Statistical Demography Konrad-Zuse-Straße 1 18057 Rostock - Germany Phone: +49 (0)381 2081 172 Fax: +49 (0)381 2081 472 cama...@demogr.mpg.de http://www.demogr.mpg.de/en/staff/camarda/default.htm - -- This mail has been sent through the MPI for Demographic ...{{dropped:10}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.spss, locale and encodings
On Wed, Apr 08, 2009 at 04:17:51PM +0200, Peter Dalgaard wrote: > Hans Ekbrand wrote: >> Someone running foreign 8.34 that is willing to test my SPSS-file? > > Someone with an SPSS file problem willing to help test the prereleases? :-) http://sociologi.cjb.net/temp/test.sav -- Hans Ekbrand (http://sociologi.cjb.net) GPG Fingerprint: 1408 C8D5 1E7D 4C9C C27E 014F 7C2C 872A 7050 614E signature.asc Description: Digital signature __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] persp3d and rgl.viewpoint for rotating 3D plots
On 4/8/2009 11:01 AM, Camarda, Carlo Giovanni wrote: Dear R-users, within the rgl-package, I would have a question about the usage of persp3d in combination of rgl.viewpoint. I am not able to figure out how to let a 3D plot rotating around likewise the example in ?rgl.viewpoint. It seems that when I use persp3d(...) I see something on my screen, which is different from what I get when it's rotating. Is there any different behavior between shade3d and persp3d? Please find below a simple example. Actually, my final aim is to save each of the rotating graphs for creating a "clip". I commented below what I would use later on, but is there any way to do it better? Take a look at ?movie3d. I think it does what you want. Duncan Murdoch Thanks a lot for your help, Carlo Giovanni Camarda library(rgl) # simplified version from help(rgl.viewpoint) shade3d(oh3d()) coo <- 1:360 for(i in 1:length(coo)) { rgl.viewpoint(coo[i]) #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") #rgl.snapshot(filename) } # simple persp3d adaption f <- function(x,y) { r <- sqrt(x^2+y^2); 10 * sin(r)/r } x <- seq(-10, 10, length= 30); y <- x z <- outer(x, y, f) persp3d(x, y, z) coo <- 1:360 for(i in 1:length(coo)) { rgl.viewpoint(coo[i]) #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") #rgl.snapshot(filename) } - Dr. Carlo Giovanni Camarda Research Scientist Max Planck Institute for Demographic Research Laboratory of Statistical Demography Konrad-Zuse-Straße 1 18057 Rostock - Germany Phone: +49 (0)381 2081 172 Fax: +49 (0)381 2081 472 cama...@demogr.mpg.de http://www.demogr.mpg.de/en/staff/camarda/default.htm - -- This mail has been sent through the MPI for Demographic ...{{dropped:10}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Is a point into an ellipse
Hi, I drew an ellipse with the package ellipse. Now I would like to know if a point is inside the ellipse. Is any R functions to do it without computing the equation of the ellipse manually? Thanks. For example, if I do "plot(ellipse(0.8), type = 'l')", I would like to know if (0,1) belongs to the drawn ellipse. Regards, Alain -- Alain Guillet Statistician and Computer Scientist SMCS - Institut de statistique - Université catholique de Louvain Bureau d.126 Voie du Roman Pays, 20 B-1348 Louvain-la-Neuve Belgium tel: +32 10 47 30 50 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] predict "interval" for lmRob?
lm's "predict" function offers an "interval" parameter to choose between 'confidence' and 'prediction' bands. In the package "robust" and for "lmRob", there is also a "predict" but it lacks such a parameter, and the documented "type" parameter has only "response" offerred. Is there some way of obtaining prediction bands from lmRob? Is there an alternative robust (linear) regression package that offers such a capability? Thanks for any and all help. - Jan Galkowski, Akamai Technologies, Cambridge, MA. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] [OT ?] rant (was : Re: Conversions From standard to metric units)
At 00:48 04/04/2009, Duncan Murdoch wrote: On 03/04/2009 5:37 PM, Emmanuel Charpentier wrote: Le vendredi 03 avril 2009 à 14:17 -0400, stephen sefick a écrit : I am starting to use R for almost any sort of calculation that I need. I am a biologist that works in the states, and there is often a need to convert from standard units to metric units. US/Imperial units are *not* standard units. But they are fun: you should see the arguments you can have about whether imperial fluid ounces are the same volume as US fluid ounces. (They're not: US ounces are bigger. But not big enough so that their gallons catch up!) Even late in the day I cannot resist sharing the astonishment of a colleague from the United States when I told him I had been taught as a child the rhyme A pint of pure water Weighs a pound and a quarter whereas he had been taught A pint's a pound The world around Duncan Murdoch The former "metric system" is now called "Système International" (International System) for a reason, which is *not* gallocentrism of a "few" 6e7 frogs, but rather laziness of about 5.6e9 losers who refuse to load their memories with meaningless conversion factors... Emmanuel Charpentier who has served his time with pounds per cubic feet, furlongs per fortnight, BTU and other figments of British/American sadistic imagination, thank you very much... # Again, didn't work the first time... Is there a package in R for this already? If not I believe that I am going to write some of the most often used in function form. My question is should I include this in my StreamMetabolism package. It is not along the same theme lines, but could loosely fit. The reason that I ask is that I don't want to clutter CRAN with a small package containing some conversion functions because I am to lazy to source them into R every time that I use them, but I also don't want the StreamMetabolism package to turn into StephenMisc Fuctions. Thoughts, comments, or suggestions would be appreciated. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Michael Dewey http://www.aghmed.fsnet.co.uk __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] vectors on top of contours, and lattice
OK, I needed to plot a set of vectors on top of a contour plot. I figured out a way to do this. I create a panel function that calls "larrows()" with arguments constructed from my vector data. Then, when I go to do the contour plot, I call contourplot() with the "panel" argument set to point to my newly created panel function. So far, so good. Now, I need to do a series of contour plots, each overlaid with a different set of vectors. I can call contourplot(z~x*y,...) but now I'm stuck, because each contour plot will be overlaid with the *same* vector plot. This won't work because each set of contoured data has its own particular set of associated vector data. (BTW, I am not doing gradients here. The vector data is its own entity, not derivable from the contoured data.) Can I somehow define a whole series of panel functions and coax contourplot() to choose the right one for each of its plots? Or do I have to revise my entire approach here? Thanks! --Sam [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Howto Plot With Transparent Background
Eik Vettorazzi wrote, On 08.04.2009 15:08: By default pdfs have a transparent background, see ?pdf and there the bg part, which can be proven by a minimal R+ LaTeX example #R code pdf("test.pdf") plot(1,1) dev.off() #minimal tex example \documentclass[a4paper,12pt]{article} \usepackage[svgnames]{xcolor} \usepackage{graphicx} \begin{document} \pagecolor{green} I really like those bizarre green pages. Thanks for sharing. :-) Christof -- Christof Winter Bioinformatics Group Biotechnologisches Zentrum Technische Universität Dresden Tatzberg 47-51 01307 Dresden Germany __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] persp3d and rgl.viewpoint for rotating 3D plots
Dear Duncan, thanks for your prompt reply. I was tempted to use movie3d, but below I copied what I get on my PC. Then, I have ImageMagick installed and, though I set convert equal to FALSE, I was not able to find where the function writes the .png files. Maybe I lack experience in managing such complex functions, but I must confess that I have quite some hard time to grasp info from help(movie3d). Thanks in advance for any help, Carlo Giovanni > library(rgl) > # set data > f <- function(x,y) { r <- sqrt(x^2+y^2); 10 * sin(r)/r } > x <- seq(-10, 10, length= 30); y <- x > z <- outer(x, y, f) > # set working directory > setwd("U:\\movie") > dir() # check character(0) > # plot > persp3d(x, y, z) > movie3d(spin3d(), duration=5, convert=TRUE) Writing movie000.png [...] Writing movie050.png Error in movie3d(spin3d(), duration = 5, convert = TRUE) : ImageMagick not found > dir() # check again character(0) > movie3d(spin3d(), duration=5, convert=FALSE) Writing movie000.png [...] Writing movie050.png > dir() # check again ?? character(0) > > version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major 2 minor 8.1 year 2008 month 12 day22 svn rev47281 language R version.string R version 2.8.1 (2008-12-22) -Original Message- From: Duncan Murdoch [mailto:murd...@stats.uwo.ca] Sent: Wednesday, April 08, 2009 5:13 PM To: Camarda, Carlo Giovanni Cc: r-h...@stat.math.ethz.ch Subject: Re: [R] persp3d and rgl.viewpoint for rotating 3D plots On 4/8/2009 11:01 AM, Camarda, Carlo Giovanni wrote: > Dear R-users, > > within the rgl-package, I would have a question about the usage of > persp3d in combination of rgl.viewpoint. > I am not able to figure out how to let a 3D plot rotating around likewise the > example in ?rgl.viewpoint. It seems that when I use persp3d(...) I see > something on my screen, which is different from what I get when it's > rotating. Is there any different behavior between shade3d and persp3d? > Please find below a simple example. > > Actually, my final aim is to save each of the rotating graphs for creating a > "clip". I commented below what I would use later on, but is there any way to > do it better? Take a look at ?movie3d. I think it does what you want. Duncan Murdoch > > Thanks a lot for your help, > Carlo Giovanni Camarda > > > library(rgl) > # simplified version from help(rgl.viewpoint) > shade3d(oh3d()) > coo <- 1:360 > for(i in 1:length(coo)) { > rgl.viewpoint(coo[i]) > #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") > #rgl.snapshot(filename) > } > > # simple persp3d adaption > f <- function(x,y) { r <- sqrt(x^2+y^2); 10 * sin(r)/r } > x <- seq(-10, 10, length= 30); y <- x > z <- outer(x, y, f) > persp3d(x, y, z) > coo <- 1:360 > for(i in 1:length(coo)) { > rgl.viewpoint(coo[i]) > #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") > #rgl.snapshot(filename) > } > > > > - > Dr. Carlo Giovanni Camarda > Research Scientist > Max Planck Institute for Demographic Research > Laboratory of Statistical Demography > Konrad-Zuse-Straße 1 > 18057 Rostock - Germany > Phone: +49 (0)381 2081 172 > Fax: +49 (0)381 2081 472 > cama...@demogr.mpg.de > http://www.demogr.mpg.de/en/staff/camarda/default.htm > - > > > -- > This mail has been sent through the MPI for Demographic ...{{dropped:10}} > > > > > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. -- This mail has been sent through the MPI for Demographic Research. Should you receive a mail that is apparently from a MPI user without this text displayed, then the address has most likely been faked. If you are uncertain about the validity of this message, please check the mail header or ask your system administrator for assistance. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Convert data frame containing time stamps to time series
Converting dates is getting stranger still. I am coercing a data frame into a ts as follows: tst1<-as.POSIXct("1/21/09 5:01",format="%m/%d/%y %H:%M") tst2<-as.POSIXct("1/28/09 3:40",format="%m/%d/%y %H:%M") tsdat<-as.ts(dat,start=tst1,end=tst2,frequency=1) This generates a ts object. But strangely enough the first column of that matrix starts at the numeric value of 841 counts up to 1139 and then starts at 1 again, only to count up from there. The restart at 1 occurs at the first day "1/21/09" at 10:00:00. What is so special about that time? This phenomenon happens several times in the long file. But the restart count is always a different number. This creates a ramp with some bumps. Can anybody explain this? Thanks in advance, Alex van der Spek > I read records using scan: > > dat<-data.frame(scan(file="KDA.csv",what=list(t="%m/%d/%y > %H:%M",f=0,p=0,d=0,o=0,s=0,a=0,l=0,c=0),skip=2,sep=",",nmax=np,flush=TRUE,na.strings=c("I/OTimeout","ArcOff-line"))) > > which results in: > >> dat[1:5,] > t fp d o sa l c > 1 1/21/09 5:01 16151 8.2 76 30 282 1060 53 7 > 2 1/21/09 5:02 16256 8.3 76 23 282 1059 54 7 > 3 1/21/09 5:03 16150 8.4 76 26 282 1059 55 7 > 4 1/21/09 5:04 16150 9.0 76 25 282 1051 57 6 > 5 1/21/09 5:05 15543 10.4 76 7 282 1024 58 6 > > I have been unable to find a way to convert this into a time series. I did > read the manuals and came across a way to coerce a data frame to a ts > object: as.ts() > > Trouble is I do not know how to keep the timestamps in column t in the > data frame above. The t column is not strings. If I do: > > plot.ts(dat) > > I can see how the first graphics panel is indeed numbers not text. So I > think scan converted the text correctly per the format string I put in. > > Much more difficult still. The datafiles I have contain invalid data, > missing values and other none relevant information. I filter this out > using subset which works brilliantly. However, how can I filter using > subset and convert to a time series afterwards. Since after subsetting > there will be 'holes' i.e. missing records. Can a ts object deal with > missing records? If so, how? Just point me to a document. I can and will > put in the work to figure it out myself. > > Thank you! > Alex van der Spek > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] persp3d and rgl.viewpoint for rotating 3D plots
On 4/8/2009 11:47 AM, Camarda, Carlo Giovanni wrote: Dear Duncan, thanks for your prompt reply. I was tempted to use movie3d, but below I copied what I get on my PC. Then, I have ImageMagick installed and, though I set convert equal to FALSE, I was not able to find where the function writes the .png files. Maybe I lack experience in managing such complex functions, but I must confess that I have quite some hard time to grasp info from help(movie3d). Presumably ImageMagick is not on your PATH, which is why movie3d() failed to find it. It writes the image files to the directory given by the dir parameter, which defaults to tempdir(). tempdir() returns a session-specific temporary directory, which R will try to clear at the end of a session. Duncan Murdoch Thanks in advance for any help, Carlo Giovanni library(rgl) # set data f <- function(x,y) { r <- sqrt(x^2+y^2); 10 * sin(r)/r } x <- seq(-10, 10, length= 30); y <- x z <- outer(x, y, f) # set working directory setwd("U:\\movie") dir() # check character(0) # plot persp3d(x, y, z) movie3d(spin3d(), duration=5, convert=TRUE) Writing movie000.png [...] Writing movie050.png Error in movie3d(spin3d(), duration = 5, convert = TRUE) : ImageMagick not found dir() # check again character(0) movie3d(spin3d(), duration=5, convert=FALSE) Writing movie000.png [...] Writing movie050.png dir() # check again ?? character(0) version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major 2 minor 8.1 year 2008 month 12 day22 svn rev47281 language R version.string R version 2.8.1 (2008-12-22) -Original Message- From: Duncan Murdoch [mailto:murd...@stats.uwo.ca] Sent: Wednesday, April 08, 2009 5:13 PM To: Camarda, Carlo Giovanni Cc: r-h...@stat.math.ethz.ch Subject: Re: [R] persp3d and rgl.viewpoint for rotating 3D plots On 4/8/2009 11:01 AM, Camarda, Carlo Giovanni wrote: Dear R-users, within the rgl-package, I would have a question about the usage of persp3d in combination of rgl.viewpoint. I am not able to figure out how to let a 3D plot rotating around likewise the example in ?rgl.viewpoint. It seems that when I use persp3d(...) I see something on my screen, which is different from what I get when it's rotating. Is there any different behavior between shade3d and persp3d? Please find below a simple example. Actually, my final aim is to save each of the rotating graphs for creating a "clip". I commented below what I would use later on, but is there any way to do it better? Take a look at ?movie3d. I think it does what you want. Duncan Murdoch Thanks a lot for your help, Carlo Giovanni Camarda library(rgl) # simplified version from help(rgl.viewpoint) shade3d(oh3d()) coo <- 1:360 for(i in 1:length(coo)) { rgl.viewpoint(coo[i]) #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") #rgl.snapshot(filename) } # simple persp3d adaption f <- function(x,y) { r <- sqrt(x^2+y^2); 10 * sin(r)/r } x <- seq(-10, 10, length= 30); y <- x z <- outer(x, y, f) persp3d(x, y, z) coo <- 1:360 for(i in 1:length(coo)) { rgl.viewpoint(coo[i]) #filename <- paste("pic", formatC(i,digits=1,flag="00"), ".png", sep="") #rgl.snapshot(filename) } - Dr. Carlo Giovanni Camarda Research Scientist Max Planck Institute for Demographic Research Laboratory of Statistical Demography Konrad-Zuse-Straße 1 18057 Rostock - Germany Phone: +49 (0)381 2081 172 Fax: +49 (0)381 2081 472 cama...@demogr.mpg.de http://www.demogr.mpg.de/en/staff/camarda/default.htm - -- This mail has been sent through the MPI for Demographic ...{{dropped:10}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- This mail has been sent through the MPI for Demographic Research. Should you receive a mail that is apparently from a MPI user without this text displayed, then the address has most likely been faked. If you are uncertain about the validity of this message, please check the mail header or ask your system administrator for assistance. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/lis
[R] Col Names
Hello, I am working with S-plus TimeSeries Object. I wonder how can I change the column names of the variable instead of using the one default? i.e to change "DEV.MSCI" to other name PositionsDEV.MSCI 04/30/1980 00:00:00.000 -0.0150328542 05/31/1980 00:00:00.000 0.0005087752 06/30/1980 00:00:00.000 0.0586794492 07/31/1980 00:00:00.000 0.0458505592 08/31/1980 00:00:00.000 0.0350926728 -- View this message in context: http://www.nabble.com/Col-Names-tp22952055p22952055.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] problem with creating a netcdf file under script sh
Hi everyone, I try to make a netcdf file which disposes a difference between 2 variables of 2netcdf files the same dimension When I programmed under R, everything is ok but when I put the code under EOF of a sh script, an error occurs: "Error, passed variable has a dim that is NOT of class dim.ncdf!". My programme is: x1<-open.ncdf("file1.nc") x2<-open.ncdf("file2.nc") y1<-get.var.ncdf(x1,"a") y2<-get.var.ncdf(x2,"a") y<- y1-y2 t<-var.def.ncdf("t",units="c",dim=x1$dim,-1) nc<-create.ncdf("test.nc",vars=t) put.var.ncdf(nc,t,y) close.ncdf(nc) Thanks in advance, T.A -- View this message in context: http://www.nabble.com/problem-with-creating-a-netcdf-file-under-script-sh-tp22952125p22952125.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Reshape - strange outputs
Dear R Users, I am using the reshape package to reformat gridded data into column format using the code shown below. However, when I display the resulting object, a single column is fomed (instead of three) and all the latitude values (which should be in either column one or two) are collected at the bottom. Also, the NA values aren't removed, despite this being requested in the code. Code: # NetCDF file has been read in and is being processed... arunoff_1986_temp <- get.var.ncdf(netcdf_1036_temp, "arunoff") # Assign row and column names columnnames <- sprintf("%.2f", seq(from = -89.75, to = 89.75, length = 360)) rnames <- sprintf("%.2f", seq(from = -179.75, to = 179.75, length = 720)) colnames(arunoff_1986_temp) <- columnnames rownames(arunoff_1986_temp) <- rnames # Melt into columnar format arunoff_1986$Longitude <- rownames(arunoff_1986) # Note: If I do: arunoff_1986$Latitude <- rownames(arunoff_1986) (i.e. change it to 'Latitude', I get the following: Warning message: In arunoff_1986$Latitude <- rownames(arunoff_1986) : Coercing LHS to a list ), thus proceed using 'Longitude' where no warning is apparent. arunoff_long_1986 <- melt(arunoff_1986, id.var="Longitude", na.rm=TRUE) > dim(arunoff_long_1986) [1] 259560 2 >head(arunoff_long_1986, n=10) # This displays what looks like one single >column, but is in fact two: column entitled "L1" is empty until the end of the >file, as shown by the 'tail' command below. value L1 1 2 3 4 5 6 7 8 9 10 > tail(arunoff_long_1986, n=10) value L1 Latitude.351 85.25 Latitude Latitude.352 85.75 Latitude Latitude.353 86.25 Latitude Latitude.354 86.75 Latitude Latitude.355 87.25 Latitude Latitude.356 87.75 Latitude Latitude.357 88.25 Latitude Latitude.358 88.75 Latitude Latitude.359 89.25 Latitude Latitude.360 89.75 Latitude I'd be very grateful indeed if anyone is able to offer assistance by way of pointing out what I've done wrong. I've spent a long time working on this and trying various options, but am srill none the wiser! I'm aiming for three columns: Latitude, Longitude, 'value'. Many thanks for any help offered, Steve __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Is a point into an ellipse
On 4/8/2009 11:17 AM, Alain Guillet wrote: Hi, I drew an ellipse with the package ellipse. Now I would like to know if a point is inside the ellipse. Is any R functions to do it without computing the equation of the ellipse manually? Thanks. For example, if I do "plot(ellipse(0.8), type = 'l')", I would like to know if (0,1) belongs to the drawn ellipse. It's probably too late by then, but if you saved the result of the call to ellipse(0.8), and treat it as a polygon, there is the function point.in.polygon in the sp package, or pip in splancs, or inside in RFOC, or inside in GEOmap, etc. Duncan Murdoch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Sweave problem, with multicolumn tables from R to LaTeX
Hi there, I have been using the example provided bellow for a while, and It was working without any problem. Nevertheless, just since 2-3 days is not working, probably because I did update.packages(). I have tried to re-install the older versions of the packages Hmisc() and xtable(), but still does not work. Can you run this example, and tell me if you got the same problems? I use linux-ubuntu, and R 2.8.1 Here is my .Rnw file %%% \documentclass{article} \usepackage[margin=1in]{geometry} \usepackage{Sweave} \SweaveOpts{echo=TRUE} \title{Why is not working this? a table from R having multi columns names} \author{Myself} \date{\today} \begin{document} \maketitle <>= #Calling some packages library(Hmisc) library(xtable) #version of the packages packageDescription('Hmisc')$Version #version of Hmisc packageDescription('xtable')$Version #version of xtable #version and features of my PC and R R.version$platform R.version$version.string @ \section{The table ready to be exported} An R table as an example. <<>>= tab.coef.allmodels <- structure(c(246.8809, 153.6952, 43.6259, 1823.686, -4.0138, -53.8741, -13.9552, -699.7154, 0, -3.0084, -0.5188, -41.7249, 4.0898, 2.0342, 0.2798, 25.636, 235.5128, 160.0244, 30.0147, 1523.0723, 3.7601, -53.9238, -3.9789, -597.1446, 0, -3.2745, -0.0025, -28.7418, 4.4203, 2.1606, 0.0124, 17.9259), .Dim = c(4L, 8L), .Dimnames = list( c("NHA", "GHA", "Hdom", "VHA"), c("Intercept", "stand2", "band3", "band7", "Intercept", "stand2", "band3", "band7" ))) tab.coef.allmodels @ \section{Producing the \LaTeX\ table} \subsection{Showing a simple output table, but not being the aimed one} The following syntax produce Table \ref{tab:coefAllmod1}, which is not what I want. <>= l <- latex(tab.coef.allmodels, file="tab.coefAllmod1.tex", rowlabel="Model for", first.hline.double=FALSE, rowlabel.just="l", caption="MLR and LME parameter estimates by stand variables", caption.loc="top", label="tab:coefAllmod1",where="!h") @ \input{tab.coefAllmod1} \subsection{What I want, but not getting it right} The following syntax produce Table \ref{tab:coefAllmod2}, which contains all the information that i want, but is not giving me the correct format. <>= l <- latex(tab.coef.allmodels, file="tab.coefAllmod2.tex", rowlabel="Model for", first.hline.double=FALSE, rowlabel.just="l", cgroup=c("MLR","LME"), n.cgroup=c(4,4), caption="MLR and LME parameter estimates by stand variables", caption.loc="top", label="tab:coefAllmod2",where="!h") @ \input{tab.coefAllmod2} I have been using this syntax for a while, and was working without any problem. Nevertheless, just since 2-3 days is not working, probably because I did {\tt update.packages()}. I have tried to re-install the older versions of the packages {\tt Hmisc()} and {\tt xtable()}, but still does not work. \vspace{1in} {\Large {\bf What Am I missing here?}} \end{document} %%% The the main .tex file, tables, the ouput pdf, and the .Rnw files are in the following url, if you want to check them http://environment.yale.edu/salas/questions/shortTable/multcolTable_short.tex http://environment.yale.edu/salas/questions/shortTable/tab.coefAllmod1.tex http://environment.yale.edu/salas/questions/shortTable/tab.coefAllmod2.tex http://environment.yale.edu/salas/questions/shortTable/multcolTable_short.pdf http://environment.yale.edu/salas/questions/shortTable/multcolTable_short.Rnw thanks in advance for your help -- --- Christian Salas E-mail:christian.sa...@yale.edu PhD candidate http://environment.yale.edu/salas School of Forestry and Environmental Studies Yale University Tel: +1-(203)-432 5126 360 Prospect St Fax:+1-(203)-432 3809 New Haven, CT 06511-2189Office: Room 35, Marsh Hall USA Yale Biometrics Lab http://environment.yale.edu/biometrics __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Col Names
Is this what you are looking for: > x Positions DEV.MSCI 04/30/1980 00:00:00.000 -0.0150328542 05/31/1980 00:00:00.000 0.0005087752 06/30/1980 00:00:00.000 0.0586794492 07/31/1980 00:00:00.000 0.0458505592 08/31/1980 00:00:00.000 0.0350926728 > colnames(x)[2] <- "NewName" > x Positions NewName 04/30/1980 00:00:00.000 -0.0150328542 05/31/1980 00:00:00.000 0.0005087752 06/30/1980 00:00:00.000 0.0586794492 07/31/1980 00:00:00.000 0.0458505592 08/31/1980 00:00:00.000 0.0350926728 On Wed, Apr 8, 2009 at 10:45 AM, livia wrote: > > Hello, > > I am working with S-plus TimeSeries Object. I wonder how can I change the > column names of the variable instead of using the one default? > > i.e to change "DEV.MSCI" to other name > Positions DEV.MSCI > 04/30/1980 00:00:00.000 -0.0150328542 > 05/31/1980 00:00:00.000 0.0005087752 > 06/30/1980 00:00:00.000 0.0586794492 > 07/31/1980 00:00:00.000 0.0458505592 > 08/31/1980 00:00:00.000 0.0350926728 > > -- > View this message in context: > http://www.nabble.com/Col-Names-tp22952055p22952055.html > Sent from the R help mailing list archive at Nabble.com. > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Colour each letter of a text string in a plot
I am inserting a DNA sequence into a plot, and hope to colourize each of the four nucleotide of the DNA sequence with a unique colour i.e., A ("red"), C ("green"), G ("blue", and T ("yellow"). I use the following codes, but the DNA sequence only shows as "red" DNA <- "ACGT" plot(1, xlim = c(0,1), ylim = c(0,1), axes=F, xlab="", ylab="", main="", xaxs="i", yaxs="i") text(x=0.5, y=0.5, labels=DNA, col=c("red", "green", "blue", "yellow")) __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Genstat into R - Randomisation test
Mike Lawrence wrote: Looks like that code implements a non-exhaustive variant of the randomization test, sometimes called a permutation test. Isn't it the other way around? (Permutation tests can be exhaustive by looking at all permutations, if a randomization test did that, then it wouldn't be random.) -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - (p.dalga...@biostat.ku.dk) FAX: (+45) 35327907 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Tinn-R pdf()
On 4/8/09, Tobias Verbeke wrote: > Hi Henning, > > > > thanks for your help, with solved the problem, although i don't why, > because when using the R editor accessible via the R console i created many > many lattice plots with the code i posted, i.e. without the print() command. > > > > At the command line, R objects (including lattice plots which are > objects) get auto-printed, i.e. the print method is invoked automatically > on these objects. > > This is not the case when you write to a pdf file. That claim is not accurate. Auto-printing should not be affected by whether a pdf device is open at the moment. I'm not familiar with TINN-R, but it's possible it executes R commands through some sort of wrapper function, and that is failing to emulate auto-printing correctly. -Deepayan > > Best, > Tobias > > > > > > > > Original-Nachricht > > > Datum: Wed, 8 Apr 2009 12:29:58 +0200 > > > Von: "ONKELINX, Thierry" > > > An: "Henning Wildhagen" , r-help@r-project.org > > > Betreff: RE: [R] Tinn-R pdf() > > > > > >Dear Henning, > > > > > > You need to print() lattice plots when using a device: > > > library(lattice) > > > > > > pdf("plot1.pdf") > > > PLOT<-(xyplot, ...) > > > print(PLOT) > > > dev.off() > > > > > > So this is not due to TINN-R. > > > > > > HTH, > > > > > > Thierry > > > > > > > > > > > > > ir. Thierry Onkelinx > > > Instituut voor natuur- en bosonderzoek / Research Institute for Nature > > > and Forest > > > Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, > > > methodology and quality assurance > > > Gaverstraat 4 > > > 9500 Geraardsbergen > > > Belgium tel. + 32 54/436 185 > > > thierry.onkel...@inbo.be www.inbo.be > > > To call in the statistician after the experiment is done may be no more > > > than asking him to perform a post-mortem examination: he may be able to > > > say what the experiment died of. > > > ~ Sir Ronald Aylmer Fisher > > > > > > The plural of anecdote is not data. > > > ~ Roger Brinner > > > > > > The combination of some data and an aching desire for an answer does not > > > ensure that a reasonable answer can be extracted from a given body of > > > data. > > > ~ John Tukey > > > > > > -Oorspronkelijk bericht- > > > Van: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] > > > Namens Henning Wildhagen > > > Verzonden: woensdag 8 april 2009 11:25 > > > Aan: r-help@r-project.org > > > Onderwerp: [R] Tinn-R pdf() > > > > > > Dear R and Tinn-R users, > > > > > > i recently switched to Tinn-R and sending code to R works fine (R 2.8.1, > > > > > > Tinn-R 2.2.0.2, OS Windows XP). However, i encountered a problem when > trying to send plots to pdf files like this: > > > > > > library(lattice) > > > > > > pdf("plot1.pdf") > > > PLOT<-(xyplot, ...) > > > PLOT > > > dev.off() > > > > > > The file "plot1.pdf" is created, but it is empty. > > > If i paste the code above directly into the R console and run it, the > > > file "plot1.pdf" is created and in this case contains "PLOT". > > > > > > I guess that some settings in Tinn-R are wrong, but i have no idea > > > which. Maybe someone has a suggestion? > > > > > > Thanks, > > > > > > Henning > > > -- __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Colour each letter of a text string in a plot
Use 'text' to write out each one: plot(0, type='n', xlim=c(0,1), ylim=c(0,1)) text(seq(0,1,length=10), rep(0.5,10), LETTERS[1:10], col=1:10) On Wed, Apr 8, 2009 at 12:15 PM, Daren Tan wrote: > I am inserting a DNA sequence into a plot, and hope to colourize each > of the four nucleotide of the DNA sequence with a unique colour i.e., > A ("red"), C ("green"), G ("blue", and T ("yellow"). I use the > following codes, but the DNA sequence only shows as "red" > > > DNA <- "ACGT" > > plot(1, xlim = c(0,1), ylim = c(0,1), axes=F, xlab="", ylab="", > main="", xaxs="i", yaxs="i") > > text(x=0.5, y=0.5, labels=DNA, col=c("red", "green", "blue", "yellow")) > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] vectors on top of contours, and lattice
On 4/8/09, Cable, Samuel B Civ USAF AFMC AFRL/RVBXI wrote: > OK, I needed to plot a set of vectors on top of a contour plot. I > figured out a way to do this. I create a panel function that calls > "larrows()" with arguments constructed from my vector data. Then, when > I go to do the contour plot, I call contourplot() with the "panel" > argument set to point to my newly created panel function. > > > > So far, so good. > > > > Now, I need to do a series of contour plots, each overlaid with a > different set of vectors. I can call contourplot(z~x*y,...) but now > I'm stuck, because each contour plot will be overlaid with the *same* > vector plot. This won't work because each set of contoured data has its > own particular set of associated vector data. (BTW, I am not doing > gradients here. The vector data is its own entity, not derivable from > the contoured data.) Can I somehow define a whole series of panel > functions and coax contourplot() to choose the right one for each of its > plots? Or do I have to revise my entire approach here? See ?packet.number for one approach. This should allow you to extract the correct component of the vector data inside the panel function. -Deepayan __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Comparing Proportions Among Groups
Hi everyone, I am trying to compare proportions among groups using the logistic regression approach as follows: 1) Fit the model log(p_i/(1-p_i)) = M + G_i, where p_i is the probability of success in group i and G_i is the effect of group i, i=1,..,I. 2) Test the hypotheses: Ho: G_1 = G_2 = ... = G_I (the probability of success is the same for all groups) versus Ha: at least two G_i's are different (at least two groups have different probabilities of success) I am wondering if there any R functions available for accomplishing these tasks. Using dummy variables is one option, but is it possible to solve this problem in R without resorting to the use of dummy variables? Many thanks and kind regards, Isabella Isabella R. Ghement, Ph.D. Ghement Statistical Consulting Company 301-7031 Blundell Road, Richmond, B.C., Canada, V6Y 1J5 Tel: 604-767-1250 Fax: 604-270-3922 E-mail: isabe...@ghement.ca Web: www.ghement.ca __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] axis values on lattice log-scale plot
On 4/7/09, Don McKenzie wrote: > I'm plotting the following (stripped of inessentials) > > xyplot(sd ~ distance | > wshed,data=sdvar.df,scales=list(x=list(log=TRUE),y=list(log=TRUE))) > > sdvar.df is a data frame, sd and distance are numeric, wshed is an ordered > factor > > trying to replicate the action of log="xy" in plot() > > The plot works fine but the axis values at the ticks are in scientific > notation, > e.g. 10^1.5, 10^2.0, ... on the X axis and 10^-.5, 10^-.4, ... on the Y > axis. > > I've searched the lattice help (and I know it must be in there:-() and > the archive but can't seem to find a way to make > the axis values display as simply numeric (i.e., 10,100,1000, ... on the X > axis and proportions on the Y axis). > > Suggestions much appreciated. You need to provide your own axis annotation function. ?xscale.components.default gives you details and an example. Also, see Figure 8.4 and 8.5 in http://lmdvr.r-forge.r-project.org/ -Deepayan __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Constrained, multiple response statistics
This sounds very similar to what I've been working on, but I'm not sure without an example. My solution has been to use an optimization that normalizes inside the objective function. The betas that are provided by optim are not normalized, however since they were normalized inside the objective function, normalizing them after the fact mirrors the internal workings of the objective function. see this example: http://markmail.org/message/ze5237m6gbgvvvyf Still, after looking at several statistical packages, and considering the thoughtful responses from my post, I think that there must be a better way using existing models, so I've been looking at other packages / models. On Tue, Apr 7, 2009 at 6:10 PM, Jonathan Greenberg wrote: > R'ers: > > I was hoping I could get some direction on this. I have a dataset of the > form: > > Y1,Y2,...,YM = f(X1,X2,...,XN), where N is >>> M > > The response data (Y1,Y2,...,YM) is frequency data, such that the sum of > all Yi = 1.0. Both Xj and Yi are continuous variables. > > I'm trying to figure out the best approach(es) to solving for the model f() > -- any ideas? I could solve each Y one at a time, but the lack of > constraint worries me, and I'm pretty sure that normalizing the data > afterwards to sum to 1.0 is not going to work out properly. Thoughts? I've > never worked with multiple response statistics before, so I'm mostly trying > to get some pointers on where to begin investigating... > > --j > > -- > > Jonathan A. Greenberg, PhD > Postdoctoral Scholar > Center for Spatial Technologies and Remote Sensing (CSTARS) > University of California, Davis > One Shields Avenue > The Barn, Room 250N > Davis, CA 95616 > Cell: 415-794-5043 > AIM: jgrn307, MSN: jgrn...@hotmail.com, Gchat: jgrn307 > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] binary version of R 2.8.x
Yes, ./configure works with no error. I guess it's a problem of my environment setup. It's my first attempt to build it myself, and I have downloaded the binaries and installed it previously. --- On Wed, 4/8/09, stephen sefick wrote: > From: stephen sefick > Subject: Re: [R] binary version of R 2.8.x > To: tomkur2006-takeh...@yahoo.com > Cc: r-help@r-project.org > Date: Wednesday, April 8, 2009, 7:35 AM > ./configure works with no errors ? > I have compiled on mac os x and debian so my advice may be > circumspect, but the errors I don't know what they are. > > On Tue, Apr 7, 2009 at 10:17 PM, > wrote: > > > > > > Hello, > > > > Is there any binary version of R 2.8.x available for > x86 Solaris 10? I need it so that I can use some R library > files. I actually tried to compile the source myself, but > it failed. I don't know much about how to compile > source files on Solaris, and it will be great if there is a > binary version of R 2.8.x available. Thanks. > > > > /***error after running make***/ > > util.c: In function `Rf_mbrtowc': > > util.c:1147: parse error before `char' > > util.c:1150: `p' undeclared (first use in this > function) > > util.c:1150: (Each undeclared identifier is reported > only once > > util.c:1150: for each function it appears in.) > > util.c:1150: `q' undeclared (first use in this > function) > > util.c:1150: `err' undeclared (first use in this > function) > > make[3]: *** [util.o] Error 1 > > make[3]: Leaving directory > `/var/tmp/R-2.8.1/src/main' > > make[2]: *** [R] Error 2 > > make[2]: Leaving directory > `/var/tmp/R-2.8.1/src/main' > > make[1]: *** [R] Error 1 > > make[1]: Leaving directory `/var/tmp/R-2.8.1/src' > > make: *** [R] Error 1 > > > > __ > > R-help@r-project.org mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > > and provide commented, minimal, self-contained, > reproducible code. > > > > > > -- > Stephen Sefick > > Let's not spend our time and resources thinking about > things that are > so little or so large that all they really do for us is > puff us up and > make us feel like gods. We are mammals, and have not > exhausted the > annoying little problems of being mammals. > > -K. Mullis __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] subscript out of bounds in eqscplot problem
Dear R users, I have the following problem. Suppose I have the following toy data set: > data m1 m2 m3 m4 m5 state [1,] 1 0 1 13 23 2 [2,] 0 1 0 23 94 2 [3,] 1 0 0 45 56 1 [4,] 0 1 0 35 84 2 [5,] 1 1 0 98 37 1 [6,] 1 1 0 68 1 2 where the last column is a categorical variable representing the outcome, columns m1 to m3 are binary and columns m4 and m5 are continuous. I wish to perform lda and then plot the results so I do: sc.dat<-scale(data[,-6]) sc.dat<-cbind(sc.dat,data[,6]) try<-lda(sc.dat[,1:5],sc..dat[,6]) tr.pred<-predict(try) eqscplot(tr.pred$x) and then I get the following error (I also have collinearity which I ignore because it doesnt matter): > eqscplot(tr.pred$x) Error in eqscplot(tr.pred$x) : subscript out of bounds Why am I getting this error? What does it mean? Any insight is highly appreciated. Thanks in advance, Pierre [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] MLE for bimodal distribution
Hello everyone, I'm trying to use mle from package stats4 to fit a bi/multi-modal distribution to some data, but I have some problems with it. Here's what I'm doing (for a bimodal distribution): # Build some fake binormally distributed data, the procedure fails also with real data, so the problem isn't here data = c(rnorm(1000, 3, 0.5), rnorm(500, 5, 0.3)) # Just to check it's bimodal plot(density(data)) f = function(m, s, m2, s2, w) { -log( sum(w*dnorm(data, mean=m, sd=s)) + sum((1-w)*dnorm(data, mean=m2, sd=s2)) ) } res = mle(f, start=list("m"=3, "s"=0.5, "m2"=5, "s2"=0.35, "w"=0.6)) This gives an error: Error in optim(start, f, method = method, hessian = TRUE, ...) : non-finite finite-difference value [2] In addition: There were 50 or more warnings (use warnings() to see the first 50) And the warnings are about dnorm producing NaNs So, my questions are: 1) What does "non-finite finite-difference value" mean? My guess would be that an Inf was passed somewhere where a finite number was expected...? I'm not sure how optim works though... 2) Is the log likelihood function I wrote correct? Can I use the same type of function for 3 modes? 3) What should I do to avoid function failure? I tried by changing the parameters, but it did not work. 4) Can I put constraints to the parameters? In particular, I would like w to be between 0 and 1. 5) Is there a way to get the log-likelihood value, so that I can compare different extimations? 6) Do you have any (possibly a bit "pratically oriented") readings about MLE to suggest? Thanks in advance Nico -- View this message in context: http://www.nabble.com/MLE-for-bimodal-distribution-tp22954970p22954970.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Minimum Spanning Tree
>I am not sure what you mean. Of course you can plot it using different >layouts, e.g. with layout.reingold.tilford (after choosing the root >vertex in some way) and then it looks like a usual tree plot, but why >would that be any better? I'd like to be able to distinguish between the nodes better. For example the image that I get looks like: http://www.nabble.com/file/p22954099/mst.gif mst.gif The tree nodes are just too close together to be able to read. I think the problem might be that some distances between the nodes are *much* smaller than other edges.The graph does not need to be directed, since the distance metric does not imply a direction. The most important part is just being able to see and differentiate between the nodes, and right now they seem to be all lumped together. How can I make it more readable? Thanks, ~josh -- View this message in context: http://www.nabble.com/Minimum-Spanning-Tree-tp22934813p22954099.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Genstat into R - Randomisation test
I was taught that Fisher proposed the F-test as a computationally simpler approximation to what he called a "Randomization test", consisting of exhaustive permutations. I never looked at the original Fisher reference myself, so this may be false. However, I haven't observed a consistent nomenclature when I have seen these tests discussed, so I typically ensure to mention whether what I'm doing is exhaustive or non-exhaustive. I do see the value in your interpretation, and think it makes sense to drop "randomization" as a name (despite it's possible historical significance) and start using "exhaustive permutation test" (to contrast with "non-exhaustive permutation test"). On Wed, Apr 8, 2009 at 1:18 PM, Peter Dalgaard wrote: > Mike Lawrence wrote: >> >> Looks like that code implements a non-exhaustive variant of the >> randomization test, sometimes called a permutation test. > > Isn't it the other way around? (Permutation tests can be exhaustive by > looking at all permutations, if a randomization test did that, then it > wouldn't be random.) > > > -- > O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B > c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K > (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 > ~~ - (p.dalga...@biostat.ku.dk) FAX: (+45) 35327907 > -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University Looking to arrange a meeting? Check my public calendar: http://tinyurl.com/mikes-public-calendar ~ Certainty is folly... I think. ~ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] binary version of R 2.8.x
Does solaris have a package managment system? And do you have all of the packages for building? It looks like you do, but I am not sure. On Wed, Apr 8, 2009 at 12:42 PM, wrote: > > Yes, ./configure works with no error. > I guess it's a problem of my environment setup. It's my first attempt to > build it myself, and I have downloaded the binaries and installed it > previously. > > > --- On Wed, 4/8/09, stephen sefick wrote: > >> From: stephen sefick >> Subject: Re: [R] binary version of R 2.8.x >> To: tomkur2006-takeh...@yahoo.com >> Cc: r-help@r-project.org >> Date: Wednesday, April 8, 2009, 7:35 AM >> ./configure works with no errors ? >> I have compiled on mac os x and debian so my advice may be >> circumspect, but the errors I don't know what they are. >> >> On Tue, Apr 7, 2009 at 10:17 PM, >> wrote: >> > >> > >> > Hello, >> > >> > Is there any binary version of R 2.8.x available for >> x86 Solaris 10? I need it so that I can use some R library >> files. I actually tried to compile the source myself, but >> it failed. I don't know much about how to compile >> source files on Solaris, and it will be great if there is a >> binary version of R 2.8.x available. Thanks. >> > >> > /***error after running make***/ >> > util.c: In function `Rf_mbrtowc': >> > util.c:1147: parse error before `char' >> > util.c:1150: `p' undeclared (first use in this >> function) >> > util.c:1150: (Each undeclared identifier is reported >> only once >> > util.c:1150: for each function it appears in.) >> > util.c:1150: `q' undeclared (first use in this >> function) >> > util.c:1150: `err' undeclared (first use in this >> function) >> > make[3]: *** [util.o] Error 1 >> > make[3]: Leaving directory >> `/var/tmp/R-2.8.1/src/main' >> > make[2]: *** [R] Error 2 >> > make[2]: Leaving directory >> `/var/tmp/R-2.8.1/src/main' >> > make[1]: *** [R] Error 1 >> > make[1]: Leaving directory `/var/tmp/R-2.8.1/src' >> > make: *** [R] Error 1 >> > >> > __ >> > R-help@r-project.org mailing list >> > https://stat.ethz.ch/mailman/listinfo/r-help >> > PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> > and provide commented, minimal, self-contained, >> reproducible code. >> > >> >> >> >> -- >> Stephen Sefick >> >> Let's not spend our time and resources thinking about >> things that are >> so little or so large that all they really do for us is >> puff us up and >> make us feel like gods. We are mammals, and have not >> exhausted the >> annoying little problems of being mammals. >> >> -K. Mullis > -- Stephen Sefick Let's not spend our time and resources thinking about things that are so little or so large that all they really do for us is puff us up and make us feel like gods. We are mammals, and have not exhausted the annoying little problems of being mammals. -K. Mullis __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Convert data frame containing time stamps to time series
have you tried using zoo and then using the function as.ts() On Wed, Apr 8, 2009 at 11:56 AM, wrote: > Converting dates is getting stranger still. I am coercing a data frame > into a ts as follows: > > > tst1<-as.POSIXct("1/21/09 5:01",format="%m/%d/%y %H:%M") > tst2<-as.POSIXct("1/28/09 3:40",format="%m/%d/%y %H:%M") > tsdat<-as.ts(dat,start=tst1,end=tst2,frequency=1) > > This generates a ts object. But strangely enough the first column of that > matrix starts at the numeric value of 841 counts up to 1139 and then > starts at 1 again, only to count up from there. The restart at 1 occurs at > the first day "1/21/09" at 10:00:00. > > What is so special about that time? This phenomenon happens several times > in the long file. But the restart count is always a different number. > This creates a ramp with some bumps. > > Can anybody explain this? > Thanks in advance, > Alex van der Spek > > >> I read records using scan: >> >> dat<-data.frame(scan(file="KDA.csv",what=list(t="%m/%d/%y >> %H:%M",f=0,p=0,d=0,o=0,s=0,a=0,l=0,c=0),skip=2,sep=",",nmax=np,flush=TRUE,na.strings=c("I/OTimeout","ArcOff-line"))) >> >> which results in: >> >>> dat[1:5,] >> t f p d o s a l c >> 1 1/21/09 5:01 16151 8.2 76 30 282 1060 53 7 >> 2 1/21/09 5:02 16256 8.3 76 23 282 1059 54 7 >> 3 1/21/09 5:03 16150 8.4 76 26 282 1059 55 7 >> 4 1/21/09 5:04 16150 9.0 76 25 282 1051 57 6 >> 5 1/21/09 5:05 15543 10.4 76 7 282 1024 58 6 >> >> I have been unable to find a way to convert this into a time series. I did >> read the manuals and came across a way to coerce a data frame to a ts >> object: as.ts() >> >> Trouble is I do not know how to keep the timestamps in column t in the >> data frame above. The t column is not strings. If I do: >> >> plot.ts(dat) >> >> I can see how the first graphics panel is indeed numbers not text. So I >> think scan converted the text correctly per the format string I put in. >> >> Much more difficult still. The datafiles I have contain invalid data, >> missing values and other none relevant information. I filter this out >> using subset which works brilliantly. However, how can I filter using >> subset and convert to a time series afterwards. Since after subsetting >> there will be 'holes' i.e. missing records. Can a ts object deal with >> missing records? If so, how? Just point me to a document. I can and will >> put in the work to figure it out myself. >> >> Thank you! >> Alex van der Spek >> >> __ >> R-help@r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >> > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Stephen Sefick Let's not spend our time and resources thinking about things that are so little or so large that all they really do for us is puff us up and make us feel like gods. We are mammals, and have not exhausted the annoying little problems of being mammals. -K. Mullis __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Reshape - strange outputs
can you provide a reproducible example? On Wed, Apr 8, 2009 at 11:59 AM, Steve Murray wrote: > > Dear R Users, > > I am using the reshape package to reformat gridded data into column format > using the code shown below. However, when I display the resulting object, a > single column is fomed (instead of three) and all the latitude values (which > should be in either column one or two) are collected at the bottom. Also, the > NA values aren't removed, despite this being requested in the code. > > Code: > > > # NetCDF file has been read in and is being processed... > > arunoff_1986_temp <- get.var.ncdf(netcdf_1036_temp, "arunoff") > > > # Assign row and column names > > columnnames <- sprintf("%.2f", seq(from = -89.75, to = 89.75, length = 360)) > rnames <- sprintf("%.2f", seq(from = -179.75, to = 179.75, length = 720)) > > colnames(arunoff_1986_temp) <- columnnames > rownames(arunoff_1986_temp) <- rnames > > > # Melt into columnar format > > arunoff_1986$Longitude <- rownames(arunoff_1986) > # Note: If I do: arunoff_1986$Latitude <- rownames(arunoff_1986) (i.e. > change it to 'Latitude', I get the following: Warning message: In > arunoff_1986$Latitude <- rownames(arunoff_1986) : Coercing LHS to a list ), > thus proceed using 'Longitude' where no warning is apparent. > > > arunoff_long_1986 <- melt(arunoff_1986, id.var="Longitude", na.rm=TRUE) > > >> dim(arunoff_long_1986) > [1] 259560 2 > >>head(arunoff_long_1986, n=10) # This displays what looks like one single >>column, but is in fact two: column entitled "L1" is empty until the end of >>the file, as shown by the 'tail' command below. > value L1 > 1 > 2 > 3 > 4 > 5 > 6 > 7 > 8 > 9 > 10 > >> tail(arunoff_long_1986, n=10) > value L1 > Latitude.351 85.25 Latitude > Latitude.352 85.75 Latitude > Latitude.353 86.25 Latitude > Latitude.354 86.75 Latitude > Latitude.355 87.25 Latitude > Latitude.356 87.75 Latitude > Latitude.357 88.25 Latitude > Latitude.358 88.75 Latitude > Latitude.359 89.25 Latitude > Latitude.360 89.75 Latitude > > > > I'd be very grateful indeed if anyone is able to offer assistance by way of > pointing out what I've done wrong. I've spent a long time working on this and > trying various options, but am srill none the wiser! I'm aiming for three > columns: Latitude, Longitude, 'value'. > > Many thanks for any help offered, > > Steve > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Stephen Sefick Let's not spend our time and resources thinking about things that are so little or so large that all they really do for us is puff us up and make us feel like gods. We are mammals, and have not exhausted the annoying little problems of being mammals. -K. Mullis __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.spss, locale and encodings
Hans Ekbrand wrote: On Wed, Apr 08, 2009 at 04:17:51PM +0200, Peter Dalgaard wrote: Hans Ekbrand wrote: Someone running foreign 8.34 that is willing to test my SPSS-file? Someone with an SPSS file problem willing to help test the prereleases? :-) http://sociologi.cjb.net/temp/test.sav No joy. > read.spss("~/Desktop/downloads/test.sav", reencode = "latin1") Error in read.spss("~/Desktop/downloads/test.sav", reencode = "latin1") : error reading system-file header In addition: Warning message: In read.spss("~/Desktop/downloads/test.sav", reencode = "latin1") : ~/Desktop/downloads/test.sav: position 143: Variable name begins with invalid character (I suppose the actual culprit could be number 144 which does indeed start with an A-ring ("ÅLDKAT")) Apparently, you can work around it like this lc <- Sys.setlocale("LC_CTYPE") Sys.setlocale("LC_CTYPE", "da_DK") x <- read.spss("~/Desktop/downloads/test.sav", reencode = "latin1") Sys.setlocale("LC_CTYPE", lc) -- which doesn't strike me as particularly logical, but whatever works -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - (p.dalga...@biostat.ku.dk) FAX: (+45) 35327907 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] predict "interval" for lmRob?
Your problem is related to the theory underlying linear models (and is an example as to why it is important to understand the theory, not just know how to plug numbers into a computer). The lm function is based on theory that assumes the y variable in normally distributed with the mean of that normal based on the model and the x values. This allows the predict function for lm to create prediction intervals based on the normal distribution, the predicted mean of that distribution, the estimated standard deviation, and the uncertainty in the predicted mean. Note that if your y variable is not normally distributed, but the sample size is large enough for the Central Limit Theorem to hold, then the confidence intervals will be approximately correct, but the prediction intervals will probably not be. When you switch to a robust regression approach, the assumption is that the y variable is not normal, so a prediction interval based on the normal distribution does not make sense. To get an appropriate prediction interval you need some information on what the distribution of the y values is (conditional on the model), but most robust techniques are not based on a specific distribution, just some properties of the distribution. Without some information (or at least an assumption) on that distribution, the predict method cannot create prediction intervals. I know that this does not answer your question, but hopefully helps you to understand what is happening. Think about what your actual scientific question is, it may be that you can answer the question without prediction intervals. If you feel that you really need the prediction intervals, then you will need to do some additional background research into what distribution you think the data comes from, then you can proceed from there. Some options include fitting a model based on that distribution, simulating data from the distribution given the model estimates and the uncertainty in those estimates, quantile regression, mixture of regressions, and others. Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare greg.s...@imail.org 801.408.8111 > -Original Message- > From: r-help-boun...@r-project.org [mailto:r-help-boun...@r- > project.org] On Behalf Of Galkowski, Jan > Sent: Wednesday, April 08, 2009 9:32 AM > To: r-help@r-project.org > Subject: [R] predict "interval" for lmRob? > > lm's "predict" function offers an "interval" parameter to choose > between 'confidence' and 'prediction' bands. In the package "robust" > and for "lmRob", there is also a "predict" but it lacks such a > parameter, and the documented "type" parameter has only "response" > offerred. Is there some way of obtaining prediction bands from lmRob? > Is there an alternative robust (linear) regression package that offers > such a capability? > > Thanks for any and all help. > > - Jan Galkowski, Akamai Technologies, Cambridge, MA. > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting- > guide.html > and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Minimum Spanning Tree
On Wed, Apr 8, 2009 at 6:12 PM, jpearl01 wrote: > > >>I am not sure what you mean. Of course you can plot it using different >>layouts, e.g. with layout.reingold.tilford (after choosing the root >>vertex in some way) and then it looks like a usual tree plot, but why >>would that be any better? > > I'd like to be able to distinguish between the nodes better. For example > the image that I get looks like: > http://www.nabble.com/file/p22954099/mst.gif mst.gif > > The tree nodes are just too close together to be able to read. I think the > problem might be that some distances between the nodes are *much* smaller > than other edges. The graph does not need to be directed, since the > distance metric does not imply a direction. The most important part is just > being able to see and differentiate between the nodes, and right now they > seem to be all lumped together. How can I make it more readable? Make the graph undirected first and then choose the right plotting parameters. E.g. the following works fine for me: set.seed(2) g <- erdos.renyi.game(100, 300, type="gnm", directed=TRUE) E(g)$weight <- runif(ecount(g)) mst <- minimum.spanning.tree(g) mst <- simplify(as.undirected(mst)) lay <- layout.reingold.tilford(mst, root=which.max(degree(mst))-1) plot(mst, layout=lay, vertex.size=5, asp=FALSE, vertex.color=NA, vertex.frame.color=NA) G. > Thanks, > ~josh > > > -- > View this message in context: > http://www.nabble.com/Minimum-Spanning-Tree-tp22934813p22954099.html > Sent from the R help mailing list archive at Nabble.com. > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Gabor Csardi UNIL DGM __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Lattice Groups
I don't understand your first question, but, since no one else has responded I can answer your second question. panel.bwplot, unlike panel.xyplot doesn't use panel.superpose when groups is not NULL. In order to get an analogous result you need to specify that you want to use panel.superpose. cols <- c("Sepal.Width", "Petal.Length", "Petal.Width") stackedData <- stack(iris[, cols]) df <- data.frame(y = stackedData$values, x = rep(iris$Species, 3), which = gl(3, nrow(iris))) bwplot(y ~ x:which, data = df, groups = which, panel=panel.superpose, panel.groups = panel.bwplot) If you don't like the default colors, you can set the fill colors with par.settings like: bwplot(y ~ x:which, data = df, groups = which, panel=panel.superpose, panel.groups = panel.bwplot, par.settings=list(superpose.symbol=list(fill=2:4))) Without the groups, the fill colors are controlled like this bwplot(y~x:which, data = df, par.settings=list(box.rectangle=list(fill=2:4))) Although if you have groups, using the groups argument is probably better. Mark Lyman Message: 41 Date: Tue, 7 Apr 2009 10:50:33 +0100 From: Richard Weeks Subject: [R] Lattice Groups To: Message-ID: Content-Type: text/plain Hi all, I'm trying to achieve a few things using the lattice package but am failing miserably. I am plotting side by side box plots and using a grouping variable, e.g. cols <- c("Sepal.Width", "Petal.Length", "Petal.Width") stackedData <- stack(iris[, cols]) df <- data.frame(y = stackedData$values, x = rep(iris$Species, 3), which = gl(3, nrow(iris))) bwplot(y ~ x:which, data = df, group = which, panel.groups = panel.bwplot) My questions are 1) How am I able to retain the median spot in the boxes? 2) How can I change the fill using the par.settings argument rather than fill =1:3 say? Best wishes, Biff __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.spss, locale and encodings
On Wed, Apr 08, 2009 at 07:12:23PM +0200, Peter Dalgaard wrote: > Apparently, you can work around it like this > > lc <- Sys.setlocale("LC_CTYPE") > Sys.setlocale("LC_CTYPE", "da_DK") > x <- read.spss("~/Desktop/downloads/test.sav", reencode = "latin1") > Sys.setlocale("LC_CTYPE", lc) > > -- which doesn't strike me as particularly logical, but whatever works THANKS a lot Peter! This works perfectly! I had been struggling with this problem way too long... -- Hans Ekbrand (http://sociologi.cjb.net) GnuPG key: 1024D/7050614E Fingerprint: 1408 C8D5 1E7D 4C9C C27E 014F 7C2C 872A 7050 614E Learn about secure email at http://www.gnupg.org signature.asc Description: Digital signature __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] predict "interval" for lmRob?
Hi Greg, Thanks for your guidance. In this case, the evidence is that the primary subpopulation of the data, accounting for observes the standard statistical model in the sense that Rice uses the term. It may by all accounts be normally distributed, and a Q-Q shows a large portion of the primary subpopulation behaves that way, out to 2 theoretical quantiles. But, for the measurement ranges of interest, the complement of the "normal subpopulation", accounting for some 20% of the total two million data points, behaves in other ways, which are, as a matter of fact, poorly understood. That's not likely to change soon. The choice of a robust regression framework and of "robust" (and possibly "quantreg" as Prof Koenker suggested) was simply to automatically fit a line to the primary subpopulation, without having to make arbitrary choices as what to keep or what to discard. Also, use of any preexisting package was simply pursued as a timesaver, worksaver, and to have some conceptual framework within to proceed other than just throwing least squares at arbitrarily chosen subsets. It sounds to me like I might use the robust regression to decide what to discard and then apply standard linear "lm" to the remainder, minding the diagnostics. Should they prove favorable, I'll proceed with the result of "lm". Thanks for pointing out the limitations of "robust" and its kin for me. BTW, if "robust" does not adopt a normal model for the y variable, what's the proper interpretation of the standard errors for slope and intercept it yields? A reference? - Jan -Original Message- From: Greg Snow [mailto:greg.s...@imail.org] Sent: Wednesday, April 08, 2009 1:20 PM To: Galkowski, Jan; r-help@r-project.org Subject: RE: predict "interval" for lmRob? Your problem is related to the theory underlying linear models (and is an example as to why it is important to understand the theory, not just know how to plug numbers into a computer). The lm function is based on theory that assumes the y variable in normally distributed with the mean of that normal based on the model and the x values. This allows the predict function for lm to create prediction intervals based on the normal distribution, the predicted mean of that distribution, the estimated standard deviation, and the uncertainty in the predicted mean. Note that if your y variable is not normally distributed, but the sample size is large enough for the Central Limit Theorem to hold, then the confidence intervals will be approximately correct, but the prediction intervals will probably not be. When you switch to a robust regression approach, the assumption is that the y variable is not normal, so a prediction interval based on the normal distribution does not make sense. To get an appropriate prediction interval you need some information on what the distribution of the y values is (conditional on the model), but most robust techniques are not based on a specific distribution, just some properties of the distribution. Without some information (or at least an assumption) on that distribution, the predict method cannot create prediction intervals. I know that this does not answer your question, but hopefully helps you to understand what is happening. Think about what your actual scientific question is, it may be that you can answer the question without prediction intervals. If you feel that you really need the prediction intervals, then you will need to do some additional background research into what distribution you think the data comes from, then you can proceed from there. Some options include fitting a model based on that distribution, simulating data from the distribution given the model estimates and the uncertainty in those estimates, quantile regression, mixture of regressions, and others. Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare greg.s...@imail.org 801.408.8111 > -Original Message- > From: r-help-boun...@r-project.org [mailto:r-help-boun...@r- > project.org] On Behalf Of Galkowski, Jan > Sent: Wednesday, April 08, 2009 9:32 AM > To: r-help@r-project.org > Subject: [R] predict "interval" for lmRob? > > lm's "predict" function offers an "interval" parameter to choose > between 'confidence' and 'prediction' bands. In the package "robust" > and for "lmRob", there is also a "predict" but it lacks such a > parameter, and the documented "type" parameter has only "response" > offerred. Is there some way of obtaining prediction bands from lmRob? > Is there an alternative robust (linear) regression package that offers > such a capability? > > Thanks for any and all help. > > - Jan Galkowski, Akamai Technologies, Cambridge, MA. > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting- > guide.htm
[R] Display a very low p-value
Hello R-Help, I ran some analysis and were hit with some low Z-score. I tried to convert it to a p-value, however, it seems like the ceiling is around 1e-16. > 1-pnorm(8) [1] 6.661338e-16 > 1-pnorm(9) [1] 0 Do you have any suggestion how I can display a very low p-value in the form of scientific number format? Thank you very much. Regards, Bhoom [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] xyplot - show values of a series on graph
Hi All, I have a very simple graph: cars <- c(1, 3, 6, 4, 9) trucks <- c(2, 5, 4, 5, 12) year <- c(2004, 2005, 2006, 2007, 2008) df2<-data.frame(cars,trucks,year) xyplot(cars+trucks~year, data=df2, type="o") I need to show the values of "cars" on the graph. How can I do this? Thanks. -- View this message in context: http://www.nabble.com/xyplot---show-values-of-a-series-on-graph-tp22956986p22956986.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] help with random forest package
The source code of the whole package is available on CRAN. All packages are submitted to CRAN is source form. There's no "rule" per se that gives the final prediction, as the final prediction is the result of plural vote by all trees in the forest. You may want to look at the varUsed() and getTree() functions. Andy From: Chrysanthi A. > Hello, > > I am a phd student in Bioinformatics and I am using the Random Forest > package in order to classify my data, but I have some questions. > Is there a function in order to visualize the trees, so as to > get the rules? > Also, could you please provide me with the code of > "randomForest" function, > as I would like to see how it works. I was wondering if I can get the > classification having the most votes over all the trees in > the forest (the > final rules that will give me the final classification). > Also, is there a > possibility to get a vector with the attributes that are > being selected for > each node during the construction of each tree? I mean, that > I would like to > know the m< the M input > attributes.. Are they selected randomly? Is there a > possibility to select > the same variable in subsequent nodes? > > Thanks a lot, > > Chrysanthi. > > [[alternative HTML version deleted]] > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > Notice: This e-mail message, together with any attachme...{{dropped:12}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Display a very low p-value
?options look at scipen On Wed, Apr 8, 2009 at 2:38 PM, Bhoom Suktitipat wrote: > Hello R-Help, > > I ran some analysis and were hit with some low Z-score. I tried to convert > it to a p-value, however, it seems like the ceiling is around 1e-16. > >> 1-pnorm(8) > [1] 6.661338e-16 >> 1-pnorm(9) > [1] 0 > > Do you have any suggestion how I can display a very low p-value in the form > of scientific number format? > > Thank you very much. > > Regards, > Bhoom > > [[alternative HTML version deleted]] > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Stephen Sefick Let's not spend our time and resources thinking about things that are so little or so large that all they really do for us is puff us up and make us feel like gods. We are mammals, and have not exhausted the annoying little problems of being mammals. -K. Mullis __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Display a very low p-value
in this case you need to use the 'lower.tail' argument, e.g., pnorm(8:15, lower.tail = FALSE) I hope it helps. Best, Dimitris Bhoom Suktitipat wrote: Hello R-Help, I ran some analysis and were hit with some low Z-score. I tried to convert it to a p-value, however, it seems like the ceiling is around 1e-16. 1-pnorm(8) [1] 6.661338e-16 1-pnorm(9) [1] 0 Do you have any suggestion how I can display a very low p-value in the form of scientific number format? Thank you very much. Regards, Bhoom [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitris Rizopoulos Assistant Professor Department of Biostatistics Erasmus University Medical Center Address: PO Box 2040, 3000 CA Rotterdam, the Netherlands Tel: +31/(0)10/7043478 Fax: +31/(0)10/7043014 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.