Re: [R] input line length in Sweave
On Tue, 24 May 2005 16:23:31 -0400 (GMT-04:00), Woodrow Setzer (WS) wrote: I am having trouble in Sweave with input line lengths. For example, I may have in my input file the chunk [...] Is there a way to get Sweave to wrap long input lines better, or get it to use my own formatting of the input? I realize I can edit the output tex file, but that is impractical in my application (too many). I am using R version 2.0.1 Patched (2005-01-26) on a Linux system (so, I suppose one possible answer is that this is fixed in 2.1.0; I cannot switch right now). No, no changes in 2.1 with respect to that problem. I have plans for a solution for this problem, but won't have time for it before summer break - in 2.2 there should be a fix for it. Best, Fritz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] rotate pie chart
I think you have to get your hands dirty on this one, but it's not too hard. Here's a function pie90() which is a tiny modification of pie(). Does that do the trick? Yes, it works perfectly fine, at least for what I wanted... :) Thanks a lot, to Lars for asking, and to Paul for getting your hands dirty! Kati PS: I know one shouldn't use pie charts at all... :) but if I do so, is there a reason why they work counter-clockwise in R? Is that convention? Sorry if its a silly question, my intuition (which might very likely be horrible) just expected them to start at 12 o'clock and fill the pie clockwisely. pie90 - function (x, labels = names(x), edges = 200, radius = 0.8, density = NULL, angle = 45, col = NULL, border = NULL, lty = NULL, main = NULL, ...) { if (!is.numeric(x) || any(is.na(x) | x = 0)) stop('x' values must be positive.) if (is.null(labels)) labels - as.character(1:length(x)) x - c(0, cumsum(x)/sum(x)) dx - diff(x) plot.new() pin - par(pin) xlim - ylim - c(-1, 1) if (pin[1] pin[2]) xlim - (pin[1]/pin[2]) * xlim else ylim - (pin[2]/pin[1]) * ylim plot.window(xlim, ylim, , asp = 1) nx - length(dx) if (is.null(col)) col - if (is.null(density)) c(white, lightblue, mistyrose, lightcyan, lavender, cornsilk) else par(fg) col - rep(col, length.out = nx) border - rep(border, length.out = nx) lty - rep(lty, length.out = nx) angle - rep(angle, length.out = nx) density - rep(density, length.out = nx) for (i in 1:nx) { n - max(2, floor(edges * dx[i])) # modified line below t2p - 2 * pi * seq(x[i], x[i + 1], length = n) + pi/2 xc - c(cos(t2p), 0) * radius yc - c(sin(t2p), 0) * radius polygon(xc, yc, density = density[i], angle = angle[i], border = border[i], col = col[i], lty = lty[i]) # modified line below t2p - 2 * pi * mean(x[i + 0:1]) + pi/2 xc - cos(t2p) * radius yc - sin(t2p) * radius if (!is.na(lab - labels[i]) lab != ) { lines(c(1, 1.05) * xc, c(1, 1.05) * yc) text(1.1 * xc, 1.1 * yc, lab, xpd = TRUE, adj = ifelse(xc 0, 1, 0), ...) } } title(main = main, ...) invisible(NULL) } Paul Sean Davis wrote: You might want to look at grid graphics and gridBase. I don't know in detail how to go about what you are asking, but grid allows you to rotate plots arbitrarily. Here are a couple of links that I think are useful. http://www.stat.auckland.ac.nz/~paul/grid/grid.html http://www.stat.auckland.ac.nz/~paul/grid/doc/rotated.pdf Sean On May 24, 2005, at 10:09 AM, Lars wrote: hey, about two weeks ago i posted a question concerning the display of two piecharts on one plot. after now being able to do so, i need to rotate them. the first piece of my pie is suppose to start at 0° but at 90°. i tried several things, all failing in the end. anyone out there who has an idea? Lars __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Dr Paul Murrell Department of Statistics The University of Auckland Private Bag 92019 Auckland New Zealand 64 9 3737599 x85392 [EMAIL PROTECTED] http://www.stat.auckland.ac.nz/~paul/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Errors in Variables
I hope somebody can help. A student of mine is doing a study on Measurement Error models (errors-in-variables, total least squares, etc.). I have an old reference to a multi archive that contains leiv3: Programs for best line fitting with errors in both coordinates. (The date is October 1989, by B.D. Ripley et al.) I have done a search for something similar in R withour success. Has this been implemented in a R-package, possibly under some sort of assumptions about variances. I would lke my student to apply some regression techniques to data that fit this profile. Any help is much appreciated. (If I have not done my search more carefully - my apologies.) Thanks Jacob Jacob L van Wyk Department of Mathematics and Statistics University of Johannesburg APK P O Box 524 Auckland Park 2006 South Africa Tel: +27-11-489-3080 Fax: +27-11-489-2832 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to break an axis?
What about simply using a log scale on the y axis? I.e. plot(..., log=y) -- Bjørn-Helge Mevik __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R unable to run on Mac OS 10.4 Tiger
Dear all, Thanks for the reply, I solved the issue last night. It appears it is was a problem with quicktime library. When I installed Tiger, I dragged my Panther quicktime plugin and this was not a good idea. R did not work (whatever version) in GUI, but would work in the terminal. I restarted the machine and the R would work as GUI, with this error message (at least this one, and many times): 2005-05-24 18:42:25.919 R[316] CFLog (21): Error loading /Library/QuickTime/DivX 5.component/Contents/MacOS/DivX 5: error code 4, error number 0 (Library not loaded: /Library/Application Support/DivXNetworks/liblame3.92.dylib Referenced from: /Library/QuickTime/DivX 5.component/Contents/MacOS/DivX 5 Reason: image not found) I installed DivX stuff and I now get only this message: 2005-05-25 09:16:31.332 R[4042] CFLog (21): Error loading /Library/QuickTime/LiveType.component/Contents/MacOS/LiveType: error code 4, error number 0 (Library not loaded: /System/Library/PrivateFrameworks/LiveType.framework/Versions/A/LiveType Referenced from: /Library/QuickTime/LiveType.component/Contents/MacOS/LiveType Reason: image not found) I suppose i will have to reinstall quicktime 7 and things will be OK. But I don't understand why R does need quicktime ?? Thanks Guillaume Hi Guillaume, There is a R-SIG-Mac alias where many of these questions are being addressed. The most likely reason is that you have a .RData file around that its trying to load. It might be missing a library or trying to connect to X11. Can you check for that 1st in a terminal window (ls -lia). Finder does not show .xxx files. If you reply to me, we can take it from there or switch to R-Sig-Mac. Rob On May 24, 2005, at 9:05 AM, Guillaume Chapron wrote: Hello, I'm running a PB G4 with Mac OS 10.4.1. I have downloaded the latest version R-2.1.0a.dmg. It appears that R does not work. It launches itself, but the window never gets ready, there is written Loading R... and a small progress wheel keeps turning indefinitely. Could someone help or suggest something? THANKS !! Guillaume __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting- guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Deleting multiple rows from a matrix
Hi, i want to delete multiple rows from a matrix. I know how to delete one by x=x[,-3] # deleting 3. column Is there a quick method how to delete e.g. the 3., 5., 7. and 8. column at once? TiA gg -- --- Gottfried Gruber mailto:[EMAIL PROTECTED] www: http://gogo.sehrsupa.net __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Deleting multiple rows from a matrix
Gottfried Gruber wrote: Hi, i want to delete multiple rows from a matrix. I know how to delete one by x=x[,-3] # deleting 3. column Is there a quick method how to delete e.g. the 3., 5., 7. and 8. column at once? TiA gg x[, -c(3, 5, 7, 8)] --sundar __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] DOING CLUSTERING WITH THE EM ALGORITHM
Hello, I am looking for documentation (PDF ?) about doing clustering with the EM Algorithm on R... I'm sorry for this question which may be trivial... Thanks... Dominique __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Placing A Legend in Titlle of a Plot
Without a small example to see what you are doing it is hard to respond. There are plenty of examples in the help for legend showing placement all over the place. So I am guessing that this might help (The only thing added from the help was the par(xpd = TRUE) x - 0:64/64 y - sin(3*pi*x) plot(x, y, type=l, col=blue, main = ) points(x, y, pch=21, bg=white) par(xpd = TRUE) legend(.4,1.25, sin(c x), pch=21, pt.bg=white, lty=1, col = blue) par(xpd = FALSE) Tom -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Josue Samayoa Sent: Wednesday, 25 May 2005 3:20 PM To: r-help@stat.math.ethz.ch Subject: [R] Placing A Legend in Titlle of a Plot Hello, I am trying to place a legend in the title area of a plot. I can not seem to find a way to get legend() to do this. Nor can I use the title() function. Does anyone know a way to do this? I would greatly appreciate any advice. Thanks Josue Samayoa Center for Biomolecular Science Engineering School of Engineering Division of Physical Biological Sciences University of California Santa Cruz (831) 459-1019 [[alternative text/enriched version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Placing A Legend in Titlle of a Plot
Le 25.05.2005 09:19, Josue Samayoa a écrit : Hello, I am trying to place a legend in the title area of a plot. I can not seem to find a way to get legend() to do this. Nor can I use the title() function. Does anyone know a way to do this? I would greatly appreciate any advice. Thanks Josue Samayoa Center for Biomolecular Science Engineering School of Engineering Division of Physical Biological Sciences University of California Santa Cruz (831) 459-1019 Hello, What kind of plot ? Not easy to know what you tried if you don't tell it. read the posting guide and then give a short reproductible example of what you tried that didn't work. Romain -- visit the R Graph Gallery : http://addictedtor.free.fr/graphiques ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Deleting multiple rows from a matrix
Le 25.05.2005 09:55, Gottfried Gruber a crit : Hi, i want to delete multiple rows from a matrix. I know how to delete one by x=x[,-3] # deleting 3. column Is there a quick method how to delete e.g. the 3., 5., 7. and 8. column at once? TiA gg Same way. Only just provide a vector, as in : x = x[,-c(3,5,7)] Romain -- visit the R Graph Gallery : http://addictedtor.free.fr/graphiques ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] reading multiple files
Dave Evens wrote: Dear All, How do I read in multiple data frames or matrices in a loop, e.g. for (i in 1:n) { channel - odbcConnectExcel(filenames) file[i] - as.data.frame(sqlFetch(channel, sheet)) } I would like file[i] to be the name of the data.frame (i.e. file[1], file[2], file[3],...etc) rather than a vector. The terminology you are using re. R objects seems to be rather confused. Quite probably you want a list of data frames such as using the filenames as names for the list elements along the follwoing lines: library(RODBC) dataframes - vector(mode=list, length=length(filenames)) names(dataframes) - filenames for(i in filenames){ channel - odbcConnectExcel(i) dataframes[[i]] - sqlFetch(channel, sheet) close(channel) } Uwe Ligges Thanks in advance for any help. Dave __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] plot in a two dimension surface with more than 2 variables
Amir Safari wrote: Dear All , How it is possible to trace a plot in a two dimension surface with more than 2 variables? Do you want a surface or a point cloud? Do you want to use some sort of projection? What is more than 2? Please specify your question more precisely. Uwe Ligges ps: library( rgl) is 3D So many THANKS - [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] DOING CLUSTERING WITH THE EM ALGORITHM
Hi, Check out the docs of the mclust and FlexMix packages: http://www.stat.washington.edu/mclust/ http://www.ci.tuwien.ac.at/~leisch/FlexMix/ Xtof. Dominique Emmanuel wrote: Hello, I am looking for documentation (PDF ?) about doing clustering with the EM Algorithm on R... I'm sorry for this question which may be trivial... Thanks... Dominique __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- A Master Carpenter has many tools and is expert with most of them.If you only know how to use a hammer, every problem starts to look like a nail. Stay away from that trap. Richard B Johnson. -- Christophe Pouzat Laboratoire de Physiologie Cerebrale CNRS UMR 8118 UFR biomedicale de l'Universite Paris V 45, rue des Saints Peres 75006 PARIS France tel: +33 (0)1 42 86 38 28 fax: +33 (0)1 42 86 38 30 web: www.biomedicale.univ-paris5.fr/physcerv/C_Pouzat.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] DOING CLUSTERING WITH THE EM ALGORITHM
Dominique Emmanuel wrote: Hello, I am looking for documentation (PDF ?) about doing clustering with the EM Algorithm on R... I'm sorry for this question which may be trivial... Thanks... Dominique Your question is a bit vague, but the mclust package might be what you're looking for. From ?Mclust: Clustering via EM initialized by hierarchical clustering for parameterized Gaussian mixture models. The number of clusters and the clustering model is chosen to maximize the BIC. HTH, --sundar __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Calling R from R and specifying wait until script is finished
Don MacQueen wrote: I don't know about efficient, but here is a way that I find to be practical, with around 100 R scripts. I don't see a reason why your functions are writing character strings using cat() into files, and after that processing those files using system calls and tools (e.g. grep) which are not available on all platforms. From my point of view the R way is to save the state into R objects (e.g. a vector of status codes with one element for each R script). Then you can easily use R functions to claculate on those objects. Uwe Ligges I create a master R script (I call it Runall.r). It begins like this: ## Execute me with R --save Runall.r Runall.log hc - TRUE if (hc) sink('Runall.out') t0runall - Sys.time() cat('\n') cat('Running script Runall.r at',format(t0runall),'\n') cat('\n') msg - try(source('ae-175.r')) ; rm.trymsg(msg) msg - try(source('ae-235.r')) ; rm.trymsg(msg) msg - try(source('ae-251.r')) ; rm.trymsg(msg) msg - try(source('ae-331.r')) ; rm.trymsg(msg) msg - try(source('ae-332.r')) ; rm.trymsg(msg) msg - try(source('ae-491.r')) ; rm.trymsg(msg) msg - try(source('ae-695.r')) ; rm.trymsg(msg) msg - try(source('ae-801.r')) ; rm.trymsg(msg) ## and so on, for as many scripts as I want to run ## although I constructed the list of scripts to run by hand, it could easily be done ## as a loop, with the script names constructed from the loop index ## the script ends with: cat('\n') t1 - Sys.time() cat('[Runall.r] Elapsed time',format(t1-t0runall),attributes(t1-t0runall)$units,'\n') if (hc) { cat('Done\n') sink() system('grep failed Runall.out Runall.info') cat('\n') system('grep succeeded Runall.out Runall.info') cat('\n') cat('See files Runall.out and Runall.info\n') cat('Done\n') } The function rm.trymsg() is this: rm.trymsg - function(msg) { if (class(msg)=='try-error') { cat('',tblid,'failed =\n') return(FALSE) } if (data.class(msg)=='list' unlist(msg)[[1]]=='bad.table') { cat('',tblid,'failed = bad.table ==\n') return(FALSE) } cat('=',tblid,'succeeded =\n') TRUE } ## and the purpose of using try() and rm.trymsg() is to let the job continue if an error occurs in one of ## the scripts. Note, however, that the text strings bad.table and tblid are unique to the task I am doing, and would not ## work in general. At 8:51 PM -0400 5/21/05, Lapointe, Pierre wrote: Hello, Let's say I have 50 R scripts to run. What would be the most efficient way to run them? I thought I could do multiple Rterms in a DOS batch file: Ex: Rterm 1.R 1.txt Rterm 2.R 2.txt ... Rterm 50.R 50.txt However, I'm afraid they will all open at the same time. I know I could pause the batch file with something like: PING 1.1.1.1 -n 1 -w 6 NUL (to delay 60 seconds) But that would require that I know how long each of my scripts take. Is there an easier way? Something like calling R from R and specifying that the script has to be finished before continuing. Thanks Pierre Lapointe *** AVIS DE NON-RESPONSABILITE:\ Ce document transmis par courri...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] DOING CLUSTERING WITH THE EM ALGORITHM
Le 25.05.2005 09:53, Dominique Emmanuel a écrit : Hello, I am looking for documentation (PDF ?) about doing clustering with the EM Algorithm on R... I'm sorry for this question which may be trivial... Thanks... Dominique Hello, You don't need to shout the subject ! You are looking for the mclust package which is available on CRAN. Moreover, trying to search 'EM' on the RSiteSeach ( http://finzi.psych.upenn.edu/search.html ) leads to it (first hit) You can even type : RSiteSearch('EM') Romain -- visit the R Graph Gallery : http://addictedtor.free.fr/graphiques ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] website reference for building R packages
Suresh Krishna wrote: it is the first link if you type making packages into the google search box here: http://maths.newcastle.edu.au/~rking/R/ Yes, please look into the archives as the posting guide asks you to do! -s. Laura Holt wrote: Hi R People: A few weeks ago, someone put a link to a website for how to for building R packages. It was very nice. Writing R Extensions is an *complete* and *up to date* documentation for this task. Package management is being steadily improved. Really, I recommend to use Writing R Extensions! If you are working on Windows, you might want to look into the R Administration and Installation manual as well which descibes how to collect and set up the required tools... Uwe Ligges But of course, I have misplaced the link. Does anyone still have that, please? It was someone from the University of Chicago, I believe. Thanks in advance. Sincerely, Laura Holt mailto: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] question: corCAR1 in lme
Hello all, I am trying to use lme to examine how a response variable (Chla) changes over time in different treatments (2 Temp 2 Light levels). Within each treatment combination, there are two replicate tanks (each with unique TankID) with coral fragments in them. All tanks are subject to the same environment until Time=0, when treatments are imposed, and Chla is measured for each tank at six times, including Time 0 just as the experiment commences. The model is: Chla.1 - lme(Chla ~ Temp*Light* Time - Temp*Light, random = ~1 | TankID, method=ML) The reasoning here is that each tanks intercept (Chla at Time 0) is a random draw from a common distribution regardless of treatment, but that the trend in Chla over time may vary among treatment combinations. Based on the help files, two separate threads from the archives, and the Pinheiro and Bates nlme 3.0 manual, I became confused about which of two ways to check for a first-order temporal autocorrelation: Chla.1b - lme(Chla ~ Temp*Light* Time - Temp*Light, random = ~1 | TankID, corr = corCAR1(form = ~Time | TankID), method=ML) Chla.1c - lme(Chla ~ Temp*Light* Time - Temp*Light, random = ~1 | TankID, corr = corCAR1(form = ~1 | TankID), method=ML) Comparing these fits with inspection of plot(ACF(chla.model1),alpha=0.05) suggests to me that there are problems with both of my attempts. For the ACF plot, the correlation at lag 1 is about 0.3, and sticks out beyond the confidence limits. By contrast, the two models' correlation parameters are not negative (phi = +0.13 and ~0 respectively), and the log-likelihood values are identical to the original model, suggesting no evidence of autocorrelation. Our times are not equally spaced (they vary from 5-8 days apart), and I gather than ACF assumes they are, but my troubleshooting (summarized below) suggests to me that my problem is bigger than this. I think I have not used corCAR1 properly, and am hoping someone can point me in the right direction. Attempted troubleshooting: 1. To check whether the discrepancy between ACF and the lme fits was due entirely to the unequal spacing of measurements, I created a bogus time variable (Time2) that was equally spaced (running from 0 to 5 in steps of 1). I then re-fit all of the above models with Time2 replacing Time in the function calls, and get the same kinds of problems (phi ~ 0 in the model fits, while ACF plot suggests a negative correlation at lag 1). 2. Still using the bogus equally-spaced time variable, I replaced corCAR1 with corAR1. Now, the two different specifications of form yield identical parameter estimates and MLLs; the estimates of phi agree with those from the ACF plot; and the models actually do fit better than the equivalent model without autocorrelation. Any advice would be greatly appreciated. Sincerely, Sean Connolly [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Problem with systemfit 0.7-3 and transformed variables
The 'systemfit' function in systemfit 0.7-3 CRAN package seems to have a problem with formulas that contain transformed (eg. log) variables. If I have my data in a data frame, apparently systemfit doesn't pass the information of where the variables should be taken to the transforming function. I'm not entirely sure if this is a bug or just a limitation, I was just surprised when I attempted to estimate a model, which I'd previously estimated with OLS using 'lm', with 2SLS using 'systemfit' and it didn't accept those transformations like 'lm' does. Here's an example: this is, of course, OK: data(kmenta) demand - q ~ p + d instr - ~ d + f fit1 - systemfit(2SLS, eqns=list(demand), inst=instr, data=kmenta) But, now if I'd like to estimate a model with logarithm of p as a regressor, an error occurs: demand2 - q ~ log(p) + d fit2 - systemfit(2SLS, eqns=list(demand2), inst=instr, data=kmenta) Error in log(p) : Object p not found However, estimating the same formula with OLS using the regular 'lm' is OK: fit2.ols - lm(demand2, kmenta) Transforming an instrument causes the same error too: instr2 - ~ log(d) + f fit3 - systemfit(2SLS, eqns=list(demand), inst=instr2, data=kmenta) Error in log(d) : Object d not found One could certainly just create those transformed variables to avoid the problem, but it would be much more convenient, if it wasn't necessary, especially if several regressors are involved. version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor1.0 year 2005 month04 day 18 language R Regards, -Mikko Pakkanen __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] colors and palettes and things...
Jeff D. Hamann wrote: After trying to find if there was a color picker in the FAQs and the help, I thought I would send a post here. I was overwhelmed with all the wonderful color choices R has predefined (discovered after typing in colors()) but can't figure out what they all (by name) look like. Is there a color picker or some other method to display all those colors next to the name? I think I can put together palettes, but another question I have then regards the building of palettes (a list of variable length I can select or create myself other than the ones defined by Palette) so I can pass these colors into functions instead of having to predefine a bunch of colors myself or use the predefined colors like terrain.colors(n)? Are there groups of colors in the colors() that I can group together to make some nice palettes for drawing barplots, etc? Thanks, Jeff. I'd like to point out some packages or code snippets that might be useful for you: # base package grDevices: ?colorRamp example(colorRamp) # CRAN package colorspace: library(help = colorspace) # CRAN package RColorBrewer: library(RColorBrewer) ?brewer.pal example(brewer.pal) # see also the last two examples by Martin Maechler in # CRAN package scatterplot3d: library(scatterplot3d) ?scatterplot3d example(scatterplot3d) Uwe Ligges __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Problem with systemfit 0.7-3 and transformed variables
On Wednesday 25 May 2005 11:43, Mikko Pakkanen wrote: The 'systemfit' function in systemfit 0.7-3 CRAN package seems to have a problem with formulas that contain transformed (eg. log) variables. If I have my data in a data frame, apparently systemfit doesn't pass the information of where the variables should be taken to the transforming function. I'm not entirely sure if this is a bug or just a limitation, I was just surprised when I attempted to estimate a model, which I'd previously estimated with OLS using 'lm', with 2SLS using 'systemfit' and it didn't accept those transformations like 'lm' does. Here's an example: this is, of course, OK: data(kmenta) demand - q ~ p + d instr - ~ d + f fit1 - systemfit(2SLS, eqns=list(demand), inst=instr, data=kmenta) But, now if I'd like to estimate a model with logarithm of p as a regressor, an error occurs: demand2 - q ~ log(p) + d fit2 - systemfit(2SLS, eqns=list(demand2), inst=instr, data=kmenta) Error in log(p) : Object p not found However, estimating the same formula with OLS using the regular 'lm' is OK: fit2.ols - lm(demand2, kmenta) Transforming an instrument causes the same error too: instr2 - ~ log(d) + f fit3 - systemfit(2SLS, eqns=list(demand), inst=instr2, data=kmenta) Error in log(d) : Object d not found One could certainly just create those transformed variables to avoid the problem, but it would be much more convenient, if it wasn't necessary, especially if several regressors are involved. We did not notice this shortcoming of systemfit() so far. Unfortunately, I don't have the time in the next few days to look into the code and figure out how to enable transformed variables. I suggest that you either create transformed variables by hand or you modify the systemfit code to enable this and send us the patch. I prefer the second :-) (that's the philosophy of open-source software like R: useRs become developeRs). Best wishes, Arne version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor1.0 year 2005 month04 day 18 language R Regards, -Mikko Pakkanen __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Arne Henningsen Department of Agricultural Economics University of Kiel Olshausenstr. 40 D-24098 Kiel (Germany) Tel: +49-431-880 4445 Fax: +49-431-880 1397 [EMAIL PROTECTED] http://www.uni-kiel.de/agrarpol/ahenningsen/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] question: corCAR1 in lme
Hi Sean, As you already guessed and as also the help-page ACF.lme() indicates: The autocorrelation function is useful for investigating serial correlation models for equally spaced data. Since you don't have equally spaced time points, you should use: corCAR1(form = ~ Time | TankID), which would indicate that Time is the position variable and *not* the order of the observations (i.e., when you use form = ~ 1 | TankID). Since you assume continuous time and you also fit a random-intercepts model, I think that the correct tool to use is the semi-variogram (i.e., look at ?Variogram()) and maybe you could also consider other correlation structures such as corExp() or corGaus(). I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/16/336899 Fax: +32/16/337015 Web: http://www.med.kuleuven.ac.be/biostat/ http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm - Original Message - From: Sean Connolly [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Wednesday, May 25, 2005 11:07 AM Subject: [R] question: corCAR1 in lme Hello all, I am trying to use lme to examine how a response variable (Chla) changes over time in different treatments (2 Temp 2 Light levels). Within each treatment combination, there are two replicate tanks (each with unique TankID) with coral fragments in them. All tanks are subject to the same environment until Time=0, when treatments are imposed, and Chla is measured for each tank at six times, including Time 0 just as the experiment commences. The model is: Chla.1 - lme(Chla ~ Temp*Light* Time - Temp*Light, random = ~1 | TankID, method=ML) The reasoning here is that each tanks intercept (Chla at Time 0) is a random draw from a common distribution regardless of treatment, but that the trend in Chla over time may vary among treatment combinations. Based on the help files, two separate threads from the archives, and the Pinheiro and Bates nlme 3.0 manual, I became confused about which of two ways to check for a first-order temporal autocorrelation: Chla.1b - lme(Chla ~ Temp*Light* Time - Temp*Light, random = ~1 | TankID, corr = corCAR1(form = ~Time | TankID), method=ML) Chla.1c - lme(Chla ~ Temp*Light* Time - Temp*Light, random = ~1 | TankID, corr = corCAR1(form = ~1 | TankID), method=ML) Comparing these fits with inspection of plot(ACF(chla.model1),alpha=0.05) suggests to me that there are problems with both of my attempts. For the ACF plot, the correlation at lag 1 is about 0.3, and sticks out beyond the confidence limits. By contrast, the two models' correlation parameters are not negative (phi = +0.13 and ~0 respectively), and the log-likelihood values are identical to the original model, suggesting no evidence of autocorrelation. Our times are not equally spaced (they vary from 5-8 days apart), and I gather than ACF assumes they are, but my troubleshooting (summarized below) suggests to me that my problem is bigger than this. I think I have not used corCAR1 properly, and am hoping someone can point me in the right direction. Attempted troubleshooting: 1. To check whether the discrepancy between ACF and the lme fits was due entirely to the unequal spacing of measurements, I created a bogus time variable (Time2) that was equally spaced (running from 0 to 5 in steps of 1). I then re-fit all of the above models with Time2 replacing Time in the function calls, and get the same kinds of problems (phi ~ 0 in the model fits, while ACF plot suggests a negative correlation at lag 1). 2. Still using the bogus equally-spaced time variable, I replaced corCAR1 with corAR1. Now, the two different specifications of form yield identical parameter estimates and MLLs; the estimates of phi agree with those from the ACF plot; and the models actually do fit better than the equivalent model without autocorrelation. Any advice would be greatly appreciated. Sincerely, Sean Connolly [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] time-ordered object list
It's often useful to view objects in time order. In Splus I can do this with objects.summary(order = dataset.date) which delivers this sort of thing ... data.class storage.modeextent object.size dataset.date reference data.frame list25 x 41700 2004.09.13 15:43 x data.frame list15 x 41175 2004.09.13 15:43 lower numeric double 100 841 2004.10.05 11:10 upper numeric double 100 841 2004.10.05 11:10 barnardfunction function12 20013 2005.04.08 13:09 sim7function function 53657 2005.04.14 15:36 .Last.fixed charactercharacter 1 52 2005.04.14 15:38 runsimfunction function 63952 2005.04.14 15:38 last.dumplist list 81186 2005.05.12 11:57 How can one obtain something similar with R ? [ R 2.1.0 / windows 2000 ] cheers Bob Kinley [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] time-ordered object list
Robert Kinley wrote: It's often useful to view objects in time order. In Splus I can do this with objects.summary(order = dataset.date) which delivers this sort of thing ... data.class storage.modeextent object.size dataset.date reference data.frame list25 x 41700 2004.09.13 15:43 x data.frame list15 x 41175 2004.09.13 15:43 lower numeric double 100 841 2004.10.05 11:10 upper numeric double 100 841 2004.10.05 11:10 barnardfunction function12 20013 2005.04.08 13:09 sim7function function 53657 2005.04.14 15:36 .Last.fixed charactercharacter 1 52 2005.04.14 15:38 runsimfunction function 63952 2005.04.14 15:38 last.dumplist list 81186 2005.05.12 11:57 How can one obtain something similar with R ? [ R 2.1.0 / windows 2000 ] You cannot. Objects in R do not have any timestamp attributes. Uwe Ligges cheers Bob Kinley [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] No ~ in JGR
R2.1.0 JGR 1.2 W2k Hello all! I´ve just installed JGR on my both R-equipped computers and am very pleased with the look and functionality. Except in one, very important, way. I can´t figure out how to get the ~ sign from the keyboard to the console. Copying it from old code works fine. Using the traditional GUI works as usual. I have a Swedish keyboard layout, where ~ shares key with ¨ and ^, all reguiring a space after the hit to produce the sign on the screen. The other signs from the key works, but AltGr + ~ followed by space just gives a blank. What do I do wrong? /CG -- CG Pettersson MSci. PhD.Stud. Swedish University of Agricultural Sciences (SLU) Dep. of Ecology and Crop production sciences (EVP). http://www.slu.se/ [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] plot 3D
Is it possible to do a graphic in 3D? This is my source: but this one is on 2D and at moment variables put on other variables, so it is difiicult to differentiate them visibly. plot(corresp(data,nf=2),xlim=c(-1,1),ylim=c(-1,1)); Thanks Sabine - ils, photos et vidéos ! [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R-devel 2.1.0 dependencies for RPMS under Fedora Core 3
The R-devel RPMS has a number of dependencies on Fedora Core development tools. When the RPMS is installed, the dependencies are resolved by automatically downloading and installing the latest version of the development tools. Looking at the README file in the fc3 directory, I was surprised by this behavior. I would recommend either changing the description of the file in the README or the behavior of the RPMS. Selected text from the README follows: These RPMS are designed to run on Fedora Core 3 (Heidelberg). R-devel: This is a stub package that contains only the documentation file on Writing R Extensions (R-exts.pdf). You should install this if you wish to compile packages from source. Chuck White PS: I've only been using Linux for a couple of weeks; I've used R under Windows for more than three years. I probably won't use the RPMS again since I would prefer to use shared libraries. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] obtaining first and last record for rows with same identifier
Francisco J. Zagmutt wrote: If you want to obtain a data frame you can use the functions head and tail like: dat=data.frame(id=rep(1:5,3),num=rnorm(15), num2=rnorm(15))#Creates data frame with id last=do.call(rbind,by(dat,dat$id,tail,1))#Selects the last observation for each id first=do.call(rbind,by(dat,dat$id,head,1))#Selects the first observation for each id newdat=rbind(first,last)#Joins data newdat=newdat[order(newdat$id),]#sorts data by id Notice that rownames will give you the original row location of the observations selected I hope this helps Francisco . . . You might also look at section 4.3 of http://biostat.mc.vanderbilt.edu/twiki/pub/Main/RS/sintro.pdf -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Can simulation involving random number generation be segmented?
Dear all, Apologies for this pedantic question that only arise when there is hardware limitation. Setting: R 2.1.0 for windows xp sp2. Scenario: To generate 1000 samples using rnorm for a simulation activity. Background: The simulation activity requires so much memory resources that generating 200 samples clogs up the PF usage as indicated in the Windows Task Manager. Therefore, short of implementing the simulation on a computer with more resources, the alternative is to generate the 1000 samples in 5 separate runs, each generating 200 samples, closing the R window and re-opening between runs. Question to be addressed: To maintain consistency and ensure reproducibility of the simulation results, the 1000 samples generated in one single run should be indentical to the 5x200 samples generated on 5 separate runs. While such consistency can be ensured using set.seed() in the case of one single run, in the case where 5 separate runs are performed, can we do the following to ensure identical samples being generated? 1. In the first run, specify the seed by, say, set.seed(1) 2. At the end of the first run, store the .Random.seed by the following manner: saved.seed.1-.Random.seed 3. At the beginning of the second run, assign the saved.seed.1 to .Random.seed as follows: .Random.seed-saved.seed.1 4. At the end of the first run, store the new .Random.seed by the following manner: saved.seed.2-.Random.seed 5. At the beginning of the second run, assign the saved.seed.2 to .Random.seed as follows: .Random.seed-saved.seed.2 This is repeated until 5 runs are completed. Will the paths of random number generation be identical in these two approaches? If not, is there a way to ensure this? Apologies again for this long-winded inquiry. Thank you. Best Lin __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Linear system
Dear R-help I have a problem solving a linear system like 353a+45b+29c=79 45a+29b+3c=5 29a+3b+4c=8 Is there any way of doing this in R? Best Regards Jim -- This e-mail and any attachment may be confidential and may also be privileged. If you are not the intended recipient, please notify us immediately and then delete this e-mail and any attachment without retaining copies or disclosing the contents thereof to any other person. Thank you. -- [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] setting elements of matrix
This is, no doubt, an easy problem, but I need help. I want to set values in a matrix based on entries in a vector. Here is an example: D is an r x m matrix initilized to all zeroes. r=6, m=5 in the example. Ro - c(1, 3, 4, 4, 6) Co - c(1, 2, 3, 4, 5) I want to set D[Ro[j], Co[j] ] to 1. So, D becomes 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 A loop works but is slow, which matters since this is being done many times in a simulation. Thanks, rv [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] setting elements of matrix
Richard Valliant wrote: This is, no doubt, an easy problem, but I need help. I want to set values in a matrix based on entries in a vector. Here is an example: D is an r x m matrix initilized to all zeroes. r=6, m=5 in the example. Ro - c(1, 3, 4, 4, 6) Co - c(1, 2, 3, 4, 5) I want to set D[Ro[j], Co[j] ] to 1. So, D becomes 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 A loop works but is slow, which matters since this is being done many times in a simulation. How about: D - matrix(0, 6, 5) D[cbind(Ro, Co)] - 1 --sundar __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Linear system
Yes, solving systems of linear equations is a strength of R and is very easy. Doug Bates has a nice article in a prior version of R News showing optimal methods for doing this in R. Here is an example using your data: X - matrix(c(353,45,29,45,29,3,29,3,4), ncol=3) y - c(79,5,8) solve(crossprod(X))%*%crossprod(X,y) See Doug's article at the following link: http://cran.r-project.org/doc/Rnews/Rnews_2004-1.pdf -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jim Gustafsson Sent: Wednesday, May 25, 2005 8:31 AM To: r-help@stat.math.ethz.ch Subject: [R] Linear system Dear R-help I have a problem solving a linear system like 353a+45b+29c=79 45a+29b+3c=5 29a+3b+4c=8 Is there any way of doing this in R? Best Regards Jim -- This e-mail and any attachment may be confidential and may also be privileged. If you are not the intended recipient, please notify us immediately and then delete this e-mail and any attachment without retaining copies or disclosing the contents thereof to any other person. Thank you. -- [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Linear system
Jim Gustafsson wrote: Dear R-help I have a problem solving a linear system like 353a+45b+29c=79 45a+29b+3c=5 29a+3b+4c=8 Is there any way of doing this in R? You already said it yourself (solve): A - rbind(c(353, 45, 29), c(45, 29, 3), c(29, 3, 4)) solve(A, c(79, 5, 8)) # [1] 0.1784625 -0.1924954 0.8505186 Uwe Ligges Best Regards Jim -- This e-mail and any attachment may be confidential and may also be privileged. If you are not the intended recipient, please notify us immediately and then delete this e-mail and any attachment without retaining copies or disclosing the contents thereof to any other person. Thank you. -- [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Linear system
On Wednesday 25 May 2005 14:30, Jim Gustafsson wrote: Dear R-help I have a problem solving a linear system like 353a+45b+29c=79 45a+29b+3c=5 29a+3b+4c=8 Is there any way of doing this in R? You can write this equation system in matrix form: M * x = y with x = ( a, b, c ) M = ( 353, 45, 29, 45, 29, 3, 29, 3, 4 ) y = ( 79, 5, 8 ) you can solve the system by x = M^(-1) * y In R you can do this by R M = matrix( c( 353, 45, 29, 45, 29, 3, 29, 3, 4 ), ncol = 3, byrow = TRUE ) R y - c( 79, 5, 8 ) R x - solve( M ) %*% y or R x - solve( M, y ) Please read a basic book about linear algebra. Arne Best Regards Jim --- --- This e-mail and any attachment may be confidential and may also be privileged. If you are not the intended recipient, please notify us immediately and then delete this e-mail and any attachment without retaining copies or disclosing the contents thereof to any other person. Thank you. --- --- [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Arne Henningsen Department of Agricultural Economics University of Kiel Olshausenstr. 40 D-24098 Kiel (Germany) Tel: +49-431-880 4445 Fax: +49-431-880 1397 [EMAIL PROTECTED] http://www.uni-kiel.de/agrarpol/ahenningsen/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] plot 3D
Navarre Sabine wrote: Is it possible to do a graphic in 3D? This is my source: but this one is on 2D and at moment variables put on other variables, so it is difiicult to differentiate them visibly. It is always the question whether 3D helps, but if you think so, you might want to take a look at one of the following packages (functions): lattice (cloud) rgl (rgl.speheres) scatterplot3d (scatterplot3d) djmrgl (plot3d) - Windows only plot(corresp(data,nf=2),xlim=c(-1,1),ylim=c(-1,1)); Your code does not help us to help. What is corresp() (I guess you mean the one from package MASS)? What is data? Why is there a ;? Because of my first two questions, this code is NOT reproducible - the posting guide asks you to specify reproducible examples. Uwe Ligges Thanks Sabine - ils, photos et vidéos ! [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] time-ordered object list
As Uwe said, you can't really do that in R. S-PLUS can do that because it stores each object as a separate file on disk, thus you can get time stamp from the file. In R everything is in the global environment that can be saved to a single workspace (.RData by default). There have been some discussions quite a while ago on how time stamping objects in R can be done. You may want to search the archive. The mbvutils package has tools for organizing objects, but I don't know if it provides time stamps. Andy From: Robert Kinley It's often useful to view objects in time order. In Splus I can do this with objects.summary(order = dataset.date) which delivers this sort of thing ... data.class storage.modeextent object.size dataset.date reference data.frame list25 x 4 1700 2004.09.13 15:43 x data.frame list15 x 4 1175 2004.09.13 15:43 lower numeric double 100 841 2004.10.05 11:10 upper numeric double 100 841 2004.10.05 11:10 barnardfunction function12 20013 2005.04.08 13:09 sim7function function 5 3657 2005.04.14 15:36 .Last.fixed charactercharacter 1 52 2005.04.14 15:38 runsimfunction function 6 3952 2005.04.14 15:38 last.dumplist list 8 1186 2005.05.12 11:57 How can one obtain something similar with R ? [ R 2.1.0 / windows 2000 ] cheers Bob Kinley [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Table Help
I'm slightly confused with aggregate, will I need to loop it until it stops returning sub lists? Or would I somehow need to convert my username column into a unique list and then do FUN=sum?? Please excuse me, I was just introduced to R yesterday to calculate statistics of over 250 gigs of text we have lol. Thanks for the reply! Ben -Original Message- From: Sean Davis To: McMurtry, Benjamin G. Cc: 'r-help@stat.math.ethz.ch' Sent: 5/24/2005 4:56 PM Subject: Re: [R] Table Help see ?aggregate Sean On May 24, 2005, at 4:39 PM, McMurtry, Benjamin G. wrote: Is there easy way using a feature like bind to sum all of collumn two where column 1 is the same? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] precision problem
I have prices that I am finding difficult to compare with ==, and , due to precision. For example: the numbers should match, with '==', but they differ in the magnitude of 1e-14 due to bunch of calculations that I run on them. Programming with java, I am used to implementing a function that compares the difference between the numbers to a pre determined precision factor. This could be very slow when I have two matrices of numbers that I could otherwise compare with a simple '==', '' or '' in R. What is teh best solution for this problem? Can I control the precision of ==, and without having to reimplement the operations in a slow way? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Contingency tables from data.frames
Gabor Grothendieck wrote: On 5/24/05, Jose Claudio Faria [EMAIL PROTECTED] wrote: Dear list, I'm trying to do a set of generic functions do make contingency tables from data.frames. It is just running nice (I'm learning R), but I think it can be better. I would like to filter the data.frame, i.e, eliminate all not numeric variables. And I don't know how to make it: please, help me. Below one of the my functions ('er' is a mention to EasieR, because I'm trying to do a package for myself and the my students): #2. Tables from data.frames #2.1---er.table.df.br (User define breaks and right) er.table.df.br - function(df, breaks = c('Sturges', 'Scott', 'FD'), right = FALSE) { if (is.data.frame(df) != 'TRUE') stop('need data.frame data') dim_df - dim(df) tmpList - list() for (i in 1:dim_df[2]) { x - as.matrix(df[ ,i]) x - na.omit(x) k - switch(breaks[1], 'Sturges' = nclass.Sturges(x), 'Scott' = nclass.scott(x), 'FD' = nclass.FD(x), stop('breaks' must be 'Sturges', 'Scott' or 'FD')) tmp - range(x) classIni - tmp[1] - tmp[2]/100 classEnd - tmp[2] + tmp[2]/100 R- classEnd-classIni h- R/k # Absolut frequency f - table(cut(x, br = seq(classIni, classEnd, h), right = right)) # Relative frequency fr - f/length(x) # Relative frequency, % frP - 100*(f/length(x)) # Cumulative frequency fac - cumsum(f) # Cumulative frequency, % facP - 100*(cumsum(f/length(x))) fi - round(f, 2) fr - round(as.numeric(fr), 2) frP - round(as.numeric(frP), 2) fac - round(as.numeric(fac), 2) facP - round(as.numeric(facP),2) # Table res - data.frame(fi, fr, frP, fac, facP) names(res) - c('Class limits', 'fi', 'fr', 'fr(%)', 'fac', 'fac(%)') tmpList - c(tmpList, list(res)) } names(tmpList) - names(df) return(tmpList) } To try the function: #a) runing nice y1=rnorm(100, 10, 1) y2=rnorm(100, 58, 4) y3=rnorm(100, 500, 10) mydf=data.frame(y1, y2, y3) #tbdf=er.table.df.br (mydf, breaks = 'Sturges', right=F) #tbdf=er.table.df.br (mydf, breaks = 'Scott', right=F) tbdf=er.table.df.br (mydf, breaks = 'FD', right=F) print(tbdf) #b) One of the problems y1=rnorm(100, 10, 1) y2=rnorm(100, 58, 4) y3=rnorm(100, 500, 10) y4=rep(letters[1:10], 10) mydf=data.frame(y1, y2, y3, y4) tbdf=er.table.df.br (mydf, breaks = 'Scott', right=F) print(tbdf) Try this: sapply(my.data.frame, is.numeric) Also you might want to look up: ?match.arg ?stopifnot ?ncol ?sapply ?lapply Thanks Gabor, you suggestion solve my basic problem. I'm working is same basic (but I think useful) functions for begginiers. Below you can see the set of functions: ### # EasyeR - Package # ### # Common function--- er.make.table - function(x, classIni, classEnd, h, right) { # Absolut frequency f - table(cut(x, br = seq(classIni, classEnd, h), right = right)) # Relative frequency fr - f/length(x) # Relative frequency, % frP - 100*(f/length(x)) # Cumulative frequency fac - cumsum(f) # Cumulative frequency, % facP - 100*(cumsum(f/length(x))) fi - round(f, 2) fr - round(as.numeric(fr), 2) frP - round(as.numeric(frP), 2) fac - round(as.numeric(fac), 2) facP - round(as.numeric(facP),2) # Table res - data.frame(fi, fr, frP, fac, facP) names(res) - c('Class limits', 'fi', 'fr', 'fr(%)', 'fac', 'fac(%)') return(res) } #1. Tables from vectors #1.1---er.table.br (User define breaks and right)--- er.table.br - function(x, breaks = c('Sturges', 'Scott', 'FD'), right = FALSE) { if (is.factor(x) || mode(x) != 'numeric') stop('need numeric data') x - na.omit(x) k - switch(breaks[1], 'Sturges' = nclass.Sturges(x), 'Scott' = nclass.scott(x), 'FD' = nclass.FD(x), stop('breaks' must be 'Sturges', 'Scott' or 'FD')) tmp - range(x) classIni - tmp[1] - abs(tmp[2])/100 classEnd - tmp[2] + abs(tmp[2])/100 R- classEnd-classIni h- R/k tbl - er.make.table(x, classIni, classEnd, h, right) return(tbl) } #1.2---er.table.kr (User define the class number (k) and right)- er.table.kr - function(x, k, right = FALSE) { if (is.factor(x) || mode(x) != 'numeric') stop('need numeric data') if ((k == '') || (k == ' ')) stop('k not defined') x - na.omit(x) tmp - range(x) classIni - tmp[1] - abs(tmp[2])/100 classEnd - tmp[2] + abs(tmp[2])/100 R- classEnd-classIni h- R/k tbl - er.make.table(x, classIni, classEnd, h, right)
Re: [R] precision problem
Omar Lakkis wrote: I have prices that I am finding difficult to compare with ==, and , due to precision. For example: the numbers should match, with '==', but they differ in the magnitude of 1e-14 due to bunch of calculations that I run on them. Programming with java, I am used to implementing a function that compares the difference between the numbers to a pre determined precision factor. This could be very slow when I have two matrices of numbers that I could otherwise compare with a simple '==', '' or '' in R. What is teh best solution for this problem? Can I control the precision of ==, and without having to reimplement the operations in a slow way? The R FAQ Why doesn't R think these numbers are equal? points you to ?all.equal Uwe Ligges __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] precision problem
On Wed, 25 May 2005, Omar Lakkis wrote: ?all.equal, I think I have prices that I am finding difficult to compare with ==, and , due to precision. For example: the numbers should match, with '==', but they differ in the magnitude of 1e-14 due to bunch of calculations that I run on them. Programming with java, I am used to implementing a function that compares the difference between the numbers to a pre determined precision factor. This could be very slow when I have two matrices of numbers that I could otherwise compare with a simple '==', '' or '' in R. What is teh best solution for this problem? Can I control the precision of ==, and without having to reimplement the operations in a slow way? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] precision problem
This is a FAQ, 7.31. -Original Message- From: Omar Lakkis [mailto:[EMAIL PROTECTED] Sent: Wednesday, May 25, 2005 10:09 AM To: r-help@stat.math.ethz.ch Subject: [R] precision problem I have prices that I am finding difficult to compare with ==, and , due to precision. For example: the numbers should match, with '==', but they differ in the magnitude of 1e-14 due to bunch of calculations that I run on them. Programming with java, I am used to implementing a function that compares the difference between the numbers to a pre determined precision factor. This could be very slow when I have two matrices of numbers that I could otherwise compare with a simple '==', '' or '' in R. What is teh best solution for this problem? Can I control the precision of ==, and without having to reimplement the operations in a slow way? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Weird function call problem
Hi, I'm encountering a very odd problem with calls to anova.mlm() from within a function. Consider the following code (data.n is a matrix of numeric values): mlmfit - lm(data.n ~ 1) mlmfit0 - lm(data.n ~ 0) print(mlmfit) anova(mlmfit,mlmfit0,test=Spherical) If I run it just like this from the console, it works just fine. If, however, I call it from within a function, e.g. using fct - function(data.k) { # same code with data.n replaced by data.k } fct(data.n) It gives me fct(data.n) Call: lm(formula = data.k ~ 1) Coefficients: ... Error in anova.mlmlist(object = mlmfit, mlmfit0, test = Spherical) : Object mlmfit not found What causes mlmfit to disappear between two lines? I haven't got the slightest clue where to look for an answer... Thank you very much for your help. Bela Bauer -- Weitersagen: GMX DSL-Flatrates mit Tempo-Garantie! Ab 4,99 Euro/Monat: http://www.gmx.net/de/go/dsl __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] precision problem
Dear Omar, Perhaps I'm missing something, but why not just subtract one matrix from the other and test the difference in relation to the precision that you require for the comparison? E.g., to test near equality, something like, abs(A - B) 1e-13. I hope this helps, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Omar Lakkis Sent: Wednesday, May 25, 2005 9:09 AM To: r-help@stat.math.ethz.ch Subject: [R] precision problem I have prices that I am finding difficult to compare with ==, and , due to precision. For example: the numbers should match, with '==', but they differ in the magnitude of 1e-14 due to bunch of calculations that I run on them. Programming with java, I am used to implementing a function that compares the difference between the numbers to a pre determined precision factor. This could be very slow when I have two matrices of numbers that I could otherwise compare with a simple '==', '' or '' in R. What is teh best solution for this problem? Can I control the precision of ==, and without having to reimplement the operations in a slow way? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Ternary Plots with continuous data
Hello I have a data base consisting of soil parameters, and tree species densities. The vcd (visualizing categorical data) package includes the ternaryplot function which plots the gravitation center of 3 prameters. I'd like to find a similar function for use with continuous data rather than categorical. Does anyone know of a function suitable for this objective? Thanks Steve Friedman __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R]: Aremos TSD files
Hi I try to read Aremos files but with any succes, could you help me please. Thanks Eric Dubé Éric Dubé Économiste / Economist Groupe économie et stratégie (dép. 657) / Economic Strategy Team (Dept. 657) Financière Banque Nationale / National Bank Financial 1155 Metcalfe 5e étage / 5th floor Montréal H3B 4S9 (514) 879-2579 [EMAIL PROTECTED] *** AVIS DE NON-RESPONSABILITE: Ce document transmis par courrier electronique est destine uniquement a la personne ou a l'entite a qui il est adresse et peut contenir des renseignements confidentiels et assujettis au secret professionnel. La confidentialite et le secret professionnel demeurent malgre l'envoi de ce document a la mauvaise adresse electronique. Si vous n'etes pas le destinataire vise ou la personne chargee de remettre ce document a son destinataire, veuillez nous en informer sans delai et detruire ce document ainsi que toute copie qui en aurait ete faite.Toute distribution, reproduction ou autre utilisation de ce document est strictement interdite. De plus, le Groupe Financiere Banque Nationale et ses filiales ne peuvent pas etre tenus responsables des dommages pouvant etre causes par des virus ou des erreurs de transmission. DISCLAIMER: \ This documentation transmitted by electronic...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Table Help
On May 25, 2005, at 9:35 AM, McMurtry, Benjamin G. wrote: I'm slightly confused with aggregate, will I need to loop it until it stops returning sub lists? Or would I somehow need to convert my username column into a unique list and then do FUN=sum?? Ben, Generally, it is OK (and encouraged) to post back to the list, just so we all learn from each others' posts. Here is an example: a - data.frame(group=rep(letters[1:3],10),values=rnorm(30)) a group values 1 a -0.146140983 2 b 0.190071267 3 c 0.207708591 4 a -1.126531350 5 b -0.013761243 6 c -0.212028409 7 a 1.012215244 8 b -0.487072132 9 c 0.667999520 10 a 0.276094321 11 b 0.835611948 12 c -1.049264490 13 a 0.399050191 14 b -1.177174650 15 c -1.471944114 16 a -0.410480517 17 b 0.006008521 18 c 1.608219688 19 a -0.120729414 20 b 1.074361542 21 c 1.782826882 22 a 1.336191577 23 b 0.721519413 24 c -0.594837423 25 a 1.120628507 26 b -2.011601905 27 c -0.076464599 28 a -1.512376840 29 b 1.182703732 30 c -1.620904211 aggregate(a[,2],by=list(a[,1]),sum) Group.1 x 1 a 0.8279207 2 b 0.3206665 3 c -0.7586886 Hope that helps. Sean __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] mixed model
Hello all, I have problem with setting up random effects. I have a model: y=x1+x2+x1*x2+z1+z1*x2 where x1, x2, x1*x2 are fixed effects and z1, z1*x2 are random effects (crossed effects) I use library(nlme) 'lme' function. My question is: how I should set up random effects? I did lme(y~x1+x2+x1:x2, data=DATA, random=~z1+z1:x2, na.action='na.omit') but it did not work. Sincerely, Natalia. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Weird function call problem
Bela Bauer wrote: Hi, I'm encountering a very odd problem with calls to anova.mlm() from within a function. Consider the following code (data.n is a matrix of numeric values): mlmfit - lm(data.n ~ 1) mlmfit0 - lm(data.n ~ 0) print(mlmfit) anova(mlmfit,mlmfit0,test=Spherical) If I run it just like this from the console, it works just fine. If, however, I call it from within a function, e.g. using fct - function(data.k) { # same code with data.n replaced by data.k } fct(data.n) It gives me fct(data.n) Call: lm(formula = data.k ~ 1) Coefficients: ... Error in anova.mlmlist(object = mlmfit, mlmfit0, test = Spherical) : Object mlmfit not found What causes mlmfit to disappear between two lines? I haven't got the slightest clue where to look for an answer... Thank you very much for your help. This looks like a bug in the anova.mlm or anova.mlmlist functions. One of them is looking in the wrong place for mlmfit. Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] weighted.mean and tapply (again)
x.1 - read.table('clipboard',header=T) x.1 GROUP VALUE FREQUENCY 1 2 278 2 2 340 3 2 416 4 2 5 3 5 2 6 1 6 2 8 1 7 3 319 8 3 410 9 3 519 10 3 6 4 by(x.1, x.1$GROUP, function(x) weighted.mean(x$VALUE, x$FREQUENCY)) x.1$GROUP: 2 [1] 2.654676 --- x.1$GROUP: 3 [1] 4.153846 Jim __ James HoltmanWhat is the problem you are trying to solve? Executive Technical Consultant -- Office of Technology, Convergys [EMAIL PROTECTED] +1 (513) 723-2929 Dan Bolser [EMAIL PROTECTED]To: R mailing list r-help@stat.math.ethz.ch uk cc: Sent by: Subject: [R] weighted.mean and tapply (again) [EMAIL PROTECTED] ath.ethz.ch 05/25/2005 11:33 I read answers to questions including the words tapply and weighted.mean, but I didn't understand either the problem (data) or the solution provided. Here is my question ... dat[1:10,] GROUP VALUE FREQUENCY 1 2 278 2 2 340 3 2 416 4 2 5 3 5 2 6 1 6 2 8 1 7 3 319 8 3 410 9 3 519 1 3 6 4 For each GROUP, I would like to calculate the weighted.mean of VALUE using the FREQUENCY as the weight, so for the snippet of data shown that would be... group.2 - weighted.mean(c(2,3,4,5,6,8),c(78,40,16,3,1,1)) group.3 - weighted.mean(c(3,4,5,6),c(19,10,19,4)) cbind(rbind(2,3),rbind(group.2,group.3)) [,1] [,2] group.22 2.654676 group.33 4.153846 I would like to use tapply to automatically do this across the whole dataset (dat) - which includes lots of other distinct grouping factors, however, like I said, I couldn't understand (and therefore apply to my data) any of the other solutions I found, so any help here would be greatly appreciated! All the best, Dan. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] mixed model
I may not follow entirely here, but your random effects structure isn't correct for lme. Also, nlme cannot handle (at least well) models with crossed random effects. A better option would be to use the lmer function. Setting up the structure for the random effects in nlme would look something like: lme(y~ fixed, data, random~= z1 + z2 |ID) Where ID is a variable that contains the grouping structure of your data. In lmer, which is more appropriate for models with crossed random effects, you lmer call might be something along the lines of: lmer(y ~ fixed + (z1|ID) + (z2|ID), data) See the most recent version of R news for more info on this topic. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of NATALIA F TCHETCHERINA Sent: Wednesday, May 25, 2005 11:50 AM To: r-help@stat.math.ethz.ch Subject: [R] mixed model Hello all, I have problem with setting up random effects. I have a model: y=x1+x2+x1*x2+z1+z1*x2 where x1, x2, x1*x2 are fixed effects and z1, z1*x2 are random effects (crossed effects) I use library(nlme) 'lme' function. My question is: how I should set up random effects? I did lme(y~x1+x2+x1:x2, data=DATA, random=~z1+z1:x2, na.action='na.omit') but it did not work. Sincerely, Natalia. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] mixed model
NATALIA F TCHETCHERINA wrote: Hello all, I have problem with setting up random effects. I have a model: y=x1+x2+x1*x2+z1+z1*x2 where x1, x2, x1*x2 are fixed effects and z1, z1*x2 are random effects (crossed effects) I use library(nlme) 'lme' function. My question is: how I should set up random effects? I did lme(y~x1+x2+x1:x2, data=DATA, random=~z1+z1:x2, na.action='na.omit') but it did not work. Sincerely, Natalia. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html The answer will depend on the types of x1, x2 and z1 (i.e. whether each of them is numeric or a factor). Because you use x1:x2 I will assume that x1, x2 and z1 are all factors. In that case the formula term x1*x2 is equivalent to x1 + x2 + x1:x2 and you could write the call to lme as lme(y ~ x1*x2, data = DATA, random = ~1|z1/x1) For lmer from the lme4 package it would be lmer(y ~ x1*x2 + (1|z1) + (1|z1:x1), data = DATA) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] The error while using R2WinBUGS
Dear all, I tried to run WinBUGS model in R by using package R2WinBUGS, but I failed because of the massage : cannot calculate DIC for model in log file of WinBUGS14. In fact, the model works in WinBUGS and I can get summaries of the model, such as density, stats..., except DIC , because DIC may not be appropriate in this kind of model due to some reasons(according to the manual of WinBUGS14), and WinBUGS cannot calculate it, but this is allowed in WinBUGS. If I just ran it in WinBUGS, that would be fine, because my goal is to get the summay not DIC, but I need to run it in R, and R must have DIC. (the example as follow). I am wondering what I can do to fix it in R? for exampe: the school example: the output in R: print(schools.sim) Inference for Bugs model at c:/schools/schools.bug 3 chains, each with 1000 iterations (first 500 discarded) n.sims = 1500 iterations saved mean sd 2.5% 25% 50% 75% 97.5% Rhat n.eff theta[1] 11.1 9.1 -3.0 5.0 10.0 16.0 31.8 1.1 39 . theta[8] 8.3 8.4 -6.6 2.8 8.1 12.7 26.2 1.0 64 mu.theta 7.6 5.9 -3.0 3.7 8.0 11.0 19.5 1.1 35 . deviance 60.8 2.5 57.0 59.1 60.2 62.1 66.6 1.0 170 pD = 3 and DIC = 63.8 (using the rule, pD = var(deviance)/2) Thanks a lot, Jia [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Problem with systemfit 0.7-3 and transformed variables
We did not notice this shortcoming of systemfit() so far. Unfortunately, I don't have the time in the next few days to look into the code and figure out how to enable transformed variables. I suggest that you either create transformed variables by hand or you modify the systemfit code to enable this and send us the patch. I prefer the second :-) (that's the philosophy of open-source software like R: useRs become developeRs). Luckily, I had some time to check the code. Debugger revealed that the problems are caused by the model.frame function which is used to compile the '$data' data frame. I don't need that data frame so much, so I just substituted model.frame with model.matrix which apparently doesn't cause this error with transformed variables. However, I tuned it a bit further, so that it should still return an identical '$data' data frame, despite the modification. I've only tested this with my example and it appears to be OK. Still, I think this should be considered a quick dirty fix -there are probably better ways to do it. But, I hope it gives the idea. Here's my attempt: [EMAIL PROTECTED] R $ diff -u systemfit.R systemfit-patched.R --- systemfit.R 2004-11-26 11:17:36.0 +0200 +++ systemfit-patched.R 2005-05-25 18:55:55.568944699 +0300 @@ -624,7 +624,11 @@ Terms - terms( eqns[[i]], data = data) m$formula - Terms m - eval(m, parent.frame()) -datai - model.frame(Terms, m) +resp - model.extract(m, response) + ## using model.matrix instead of model.frame, need to get the output variable separately +datai - data.frame(cbind(resp, (model.matrix(Terms, m))[,-1])) + ## I guess there's a better way to extract the name of the output variable? + names(datai)[1] - as.character(terms(eqns[[i]]))[2] if(method==2SLS | method==3SLS) { #datai - cbind( datai, model.frame( instl[[i]] )) # the following lines have to be substituted for the previous @@ -634,7 +638,8 @@ Terms - terms(instl[[i]], data = data) m$formula - Terms m - eval(m, parent.frame()) - datai - cbind( datai, model.frame(Terms, m)) + ## used previously model.frame + datai - cbind( datai, as.data.frame((model.matrix(Terms, m))[,-1])) } if(i==1) { Regards, -Mikko. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] weighted.mean and tapply (again)
In this simple case you can just coerce directly to a vector like so: foo - by(dat, dat$GROUP, function(x) {weighted.mean(x$VALUE, x$FREQUENCY)}) bar - as.vector(foo) but look at the examples for ?by; in particular the use of 'sapply'. Norm -Original Message- From: Dan Bolser [mailto:[EMAIL PROTECTED] Sent: Wednesday, May 25, 2005 9:02 AM To: [EMAIL PROTECTED] Subject: RE: [R] weighted.mean and tapply (again) On Wed, 25 May 2005 [EMAIL PROTECTED] wrote: I think by will do what you want: by(dat, dat$GROUP, function(x) {weighted.mean(x$VALUE, x$FREQUENCY)}) Norm Thanks Norm and Jim. My question is now... How do you turn a quote list of class 'by' /quote into a vector of results (one value per GROUP)? Cheers guys! Dan. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dan Bolser Sent: Wednesday, May 25, 2005 8:34 AM To: R mailing list Subject: [R] weighted.mean and tapply (again) I read answers to questions including the words tapply and weighted.mean, but I didn't understand either the problem (data) or the solution provided. Here is my question ... dat[1:10,] GROUP VALUE FREQUENCY 1 2 278 2 2 340 3 2 416 4 2 5 3 5 2 6 1 6 2 8 1 7 3 319 8 3 410 9 3 519 1 3 6 4 For each GROUP, I would like to calculate the weighted.mean of VALUE using the FREQUENCY as the weight, so for the snippet of data shown that would be... group.2 - weighted.mean(c(2,3,4,5,6,8),c(78,40,16,3,1,1)) group.3 - weighted.mean(c(3,4,5,6),c(19,10,19,4)) cbind(rbind(2,3),rbind(group.2,group.3)) [,1] [,2] group.22 2.654676 group.33 4.153846 I would like to use tapply to automatically do this across the whole dataset (dat) - which includes lots of other distinct grouping factors, however, like I said, I couldn't understand (and therefore apply to my data) any of the other solutions I found, so any help here would be greatly appreciated! All the best, Dan. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R-SIG-Mac] Re: [R] R unable to run on Mac OS 10.4 Tiger
On May 25, 2005, at 3:19 AM, Guillaume Chapron wrote: I suppose i will have to reinstall quicktime 7 and things will be OK. But I don't understand why R does need quicktime ?? Neither R nor R.app does. Chances are that you pretty much broke your Tiger, because all R.app does is to link Cocoa which links AppKit. On my Tiger machine it doesn't even touch QuickTime, so it must be your plugin or whatever you did to poor Tiger ;). Why don't you just remove /Library/Quicktime? It's not a part of Tiger. Tiger's QuickTime files are in /System/Library. If you want to understand better who's loading what, set DYLD_PRINT_LIBRARIES when running R or R.app. Cheers, Simon __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Can simulation involving random number generation be segm ented?
I don't see why you need to shut R down between runs. Can you just replace each time you generate a new 200 samples? You might also find gc() helpful if you generate a lot of intermediate objects in a single run. Delete objects you no longer need (unless you're going to replace them and can afford to keep them around), then an explicit request gc() might help keep you under the wire. Reid Huntsinger -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Uwe Ligges Sent: Wednesday, May 25, 2005 8:56 AM To: Dr L. Y Hin Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Can simulation involving random number generation be segmented? Dr L. Y Hin wrote: Dear all, Apologies for this pedantic question that only arise when there is hardware limitation. Setting: R 2.1.0 for windows xp sp2. Scenario: To generate 1000 samples using rnorm for a simulation activity. Background: The simulation activity requires so much memory resources that generating 200 samples clogs up the PF usage as indicated in the Windows Task Manager. Therefore, short of implementing the simulation on a computer with more resources, the alternative is to generate the 1000 samples in 5 separate runs, each generating 200 samples, closing the R window and re-opening between runs. Question to be addressed: To maintain consistency and ensure reproducibility of the simulation results, the 1000 samples generated in one single run should be indentical to the 5x200 samples generated on 5 separate runs. While such consistency can be ensured using set.seed() in the case of one single run, in the case where 5 separate runs are performed, can we do the following to ensure identical samples being generated? 1. In the first run, specify the seed by, say, set.seed(1) 2. At the end of the first run, store the .Random.seed by the following manner: saved.seed.1-.Random.seed 3. At the beginning of the second run, assign the saved.seed.1 to .Random.seed as follows: .Random.seed-saved.seed.1 4. At the end of the first run, store the new .Random.seed by the following manner: saved.seed.2-.Random.seed 5. At the beginning of the second run, assign the saved.seed.2 to .Random.seed as follows: .Random.seed-saved.seed.2 This is repeated until 5 runs are completed. Will the paths of random number generation be identical in these two approaches? Yes. Uwe Ligges If not, is there a way to ensure this? Apologies again for this long-winded inquiry. Thank you. Best Lin __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] weighted.mean and tapply (again)
In this simple case you can just coerce directly to a vector like so: foo - by(dat, dat$GROUP, function(x) {weighted.mean(x$VALUE, x$FREQUENCY)}) bar - as.vector(foo) but look at the examples for ?by; in particular the use of 'sapply'. Norm As a general practice, I believe unlist() is preferable. ?unlist explains why. Cheers, Bert Gunter __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] The error while using R2WinBUGS
I don't know if I understand you correctly, but I think you missed to read ?bugs carefully enough: bugs(., DIC = FALSE) should disable calculation of DIC statistics. If your example still fails, please specify a reproducible example in a private message. Asking the package maintainer / authors would have been a better idea for such a specialized question. Uwe Ligges BTW: You might want to try out the new CRAN package BRugs, which provides an interface to the (included) OpenBUGS (the next version of WinBUGS). Currenty, it only works on Windows, though. Li, Jia wrote: Dear all, I tried to run WinBUGS model in R by using package R2WinBUGS, but I failed because of the massage : cannot calculate DIC for model in log file of WinBUGS14. In fact, the model works in WinBUGS and I can get summaries of the model, such as density, stats..., except DIC , because DIC may not be appropriate in this kind of model due to some reasons(according to the manual of WinBUGS14), and WinBUGS cannot calculate it, but this is allowed in WinBUGS. If I just ran it in WinBUGS, that would be fine, because my goal is to get the summay not DIC, but I need to run it in R, and R must have DIC. (the example as follow). I am wondering what I can do to fix it in R? for exampe: the school example: the output in R: print(schools.sim) Inference for Bugs model at c:/schools/schools.bug 3 chains, each with 1000 iterations (first 500 discarded) n.sims = 1500 iterations saved mean sd 2.5% 25% 50% 75% 97.5% Rhat n.eff theta[1] 11.1 9.1 -3.0 5.0 10.0 16.0 31.8 1.1 39 . theta[8] 8.3 8.4 -6.6 2.8 8.1 12.7 26.2 1.0 64 mu.theta 7.6 5.9 -3.0 3.7 8.0 11.0 19.5 1.1 35 . deviance 60.8 2.5 57.0 59.1 60.2 62.1 66.6 1.0 170 pD = 3 and DIC = 63.8 (using the rule, pD = var(deviance)/2) Thanks a lot, Jia [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Rounding fractional numbers to nearest fraction
Hi all, I've got a matrix of fractional data that is all positive and greater than zero that I would like to loosely classify, for lack of a better word. It looks something like this : 1.07 1.11 1.27 1.59 0.97 0.76 2.23 0.98 0.71 0.88 1.19 1.02 What I'm looking for is a way to round these numbers to the nearest 0.25, i.e. the above matrix would be transformed to : 1.00 1.00 1.25 1.50 1.00 0.75 2.25 1.00 0.75 1.00 1.25 1.00 Anyone have a clever way to do this?? Thanks in advance, Ken __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Rounding fractional numbers to nearest fraction
Multiply by 4, round and divide by 4. a - c(1.15,5.82) round(a*4,digits=0)/4 -Original Message- From: Ken Termiso [mailto:[EMAIL PROTECTED] Sent: Wednesday, May 25, 2005 1:27 PM To: r-help@stat.math.ethz.ch Subject: [R] Rounding fractional numbers to nearest fraction Hi all, I've got a matrix of fractional data that is all positive and greater than zero that I would like to loosely classify, for lack of a better word. It looks something like this : 1.07 1.11 1.27 1.59 0.97 0.76 2.23 0.98 0.71 0.88 1.19 1.02 What I'm looking for is a way to round these numbers to the nearest 0.25, i.e. the above matrix would be transformed to : 1.00 1.00 1.25 1.50 1.00 0.75 2.25 1.00 0.75 1.00 1.25 1.00 Anyone have a clever way to do this?? Thanks in advance, Ken __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Rounding fractional numbers to nearest fraction
Thx for the help... mdiv - function(dec) { dec - dec * 4 dec - round(dec, 0) dec - dec / 4 dec } where m is a matrix -- m1 - apply(m, 1, mdiv) ...but it looks like as.integer(x.1*4)/4 or (round(x.1*4,0))/4 might be easier... thx again, ken From: [EMAIL PROTECTED] To: Ken Termiso [EMAIL PROTECTED] Subject: Re: [R] Rounding fractional numbers to nearest fraction Date: Wed, 25 May 2005 14:30:44 -0400 x.1 [1] 1.07 1.11 1.27 1.59 0.97 0.76 2.23 0.98 0.71 0.88 1.19 1.02 as.integer(x.1*4)/4 [1] 1.00 1.00 1.25 1.50 0.75 0.75 2.00 0.75 0.50 0.75 1.00 1.00 Jim __ James HoltmanWhat is the problem you are trying to solve? Executive Technical Consultant -- Convergys Labs [EMAIL PROTECTED] +1 (513) 723-2929 Ken Termiso [EMAIL PROTECTED]To: r-help@stat.math.ethz.ch om cc: Sent by: Subject: [R] Rounding fractional numbers to nearest fraction [EMAIL PROTECTED] ath.ethz.ch 05/25/2005 13:27 Hi all, I've got a matrix of fractional data that is all positive and greater than zero that I would like to loosely classify, for lack of a better word. It looks something like this : 1.07 1.11 1.27 1.59 0.97 0.76 2.23 0.98 0.71 0.88 1.19 1.02 What I'm looking for is a way to round these numbers to the nearest 0.25, i.e. the above matrix would be transformed to : 1.00 1.00 1.25 1.50 1.00 0.75 2.25 1.00 0.75 1.00 1.25 1.00 Anyone have a clever way to do this?? Thanks in advance, Ken __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Rounding fractional numbers to nearest fraction
From: Ken Termiso Thx for the help... mdiv - function(dec) { dec - dec * 4 dec - round(dec, 0) dec - dec / 4 dec } where m is a matrix -- m1 - apply(m, 1, mdiv) You don't need the apply(): m - matrix(runif(6), 2, 3) m [,1] [,2] [,3] [1,] 0.6885473 0.4598892 0.2738169 [2,] 0.2393461 0.8863284 0.7891100 round(m * 4) / 4 [,1] [,2] [,3] [1,] 0.75 0.5 0.25 [2,] 0.25 1.0 0.75 ...but it looks like as.integer(x.1*4)/4 or (round(x.1*4,0))/4 might be easier... Be careful: as.integer() trucate the fractional part (so as.integer(1.999) is 1 intead of 2), which may not be what you want. Andy thx again, ken From: [EMAIL PROTECTED] To: Ken Termiso [EMAIL PROTECTED] Subject: Re: [R] Rounding fractional numbers to nearest fraction Date: Wed, 25 May 2005 14:30:44 -0400 x.1 [1] 1.07 1.11 1.27 1.59 0.97 0.76 2.23 0.98 0.71 0.88 1.19 1.02 as.integer(x.1*4)/4 [1] 1.00 1.00 1.25 1.50 0.75 0.75 2.00 0.75 0.50 0.75 1.00 1.00 Jim __ James HoltmanWhat is the problem you are trying to solve? Executive Technical Consultant -- Convergys Labs [EMAIL PROTECTED] +1 (513) 723-2929 Ken Termiso [EMAIL PROTECTED]To: r-help@stat.math.ethz.ch om cc: Sent by: Subject: [R] Rounding fractional numbers to nearest fraction [EMAIL PROTECTED] ath.ethz.ch 05/25/2005 13:27 Hi all, I've got a matrix of fractional data that is all positive and greater than zero that I would like to loosely classify, for lack of a better word. It looks something like this : 1.07 1.11 1.27 1.59 0.97 0.76 2.23 0.98 0.71 0.88 1.19 1.02 What I'm looking for is a way to round these numbers to the nearest 0.25, i.e. the above matrix would be transformed to : 1.00 1.00 1.25 1.50 1.00 0.75 2.25 1.00 0.75 1.00 1.25 1.00 Anyone have a clever way to do this?? Thanks in advance, Ken __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] cor vs cor.test
Using Windows System, R 2.1.0 d is a data frame, 48 rows, 10 columns cor(d) works properly providing all pairwise Pearson correlation coefficients among columns cor.test(d) gives error message Error in cor.test.default(d) : argument y is missing, with no default Why? Thanks, MCG __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can simulation involving random number generation be segmented?
You might consider the rlecuyer package, which provides parallel streams and avoids the potential worst case scenario from arbitrary seed setting. On 5/25/05, Dr L. Y Hin [EMAIL PROTECTED] wrote: Dear all, Apologies for this pedantic question that only arise when there is hardware limitation. Setting: R 2.1.0 for windows xp sp2. Scenario: To generate 1000 samples using rnorm for a simulation activity. Background: The simulation activity requires so much memory resources that generating 200 samples clogs up the PF usage as indicated in the Windows Task Manager. Therefore, short of implementing the simulation on a computer with more resources, the alternative is to generate the 1000 samples in 5 separate runs, each generating 200 samples, closing the R window and re-opening between runs. Question to be addressed: To maintain consistency and ensure reproducibility of the simulation results, the 1000 samples generated in one single run should be indentical to the 5x200 samples generated on 5 separate runs. While such consistency can be ensured using set.seed() in the case of one single run, in the case where 5 separate runs are performed, can we do the following to ensure identical samples being generated? 1. In the first run, specify the seed by, say, set.seed(1) 2. At the end of the first run, store the .Random.seed by the following manner: saved.seed.1-.Random.seed 3. At the beginning of the second run, assign the saved.seed.1 to .Random.seed as follows: .Random.seed-saved.seed.1 4. At the end of the first run, store the new .Random.seed by the following manner: saved.seed.2-.Random.seed 5. At the beginning of the second run, assign the saved.seed.2 to .Random.seed as follows: .Random.seed-saved.seed.2 This is repeated until 5 runs are completed. Will the paths of random number generation be identical in these two approaches? If not, is there a way to ensure this? Apologies again for this long-winded inquiry. Thank you. Best Lin __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- best, -tony Commit early,commit often, and commit in a repository from which we can easily roll-back your mistakes (AJR, 4Jan05). A.J. Rossini [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] website reference for building R packages
I used this tutorial and completed steps 1-6 without errors, but on step 7 I tried to run R CMD check on my package called 'ram' directory one level above the 'ram' folder and I got the following error? I:\R_HOMER CMD check ram * checking for working latex ...Error: environment variable TMPDIR not set (or s et to unusable value) and no default available. at c:\R\rw2010\share\perl/R/Utils.pm line 72 Can anyone help me with this? Thanks, Roger On 5/25/05, Uwe Ligges [EMAIL PROTECTED] wrote: Suresh Krishna wrote: it is the first link if you type making packages into the google search box here: http://maths.newcastle.edu.au/~rking/R/ Yes, please look into the archives as the posting guide asks you to do! -s. Laura Holt wrote: Hi R People: A few weeks ago, someone put a link to a website for how to for building R packages. It was very nice. Writing R Extensions is an *complete* and *up to date* documentation for this task. Package management is being steadily improved. Really, I recommend to use Writing R Extensions! If you are working on Windows, you might want to look into the R Administration and Installation manual as well which descibes how to collect and set up the required tools... Uwe Ligges But of course, I have misplaced the link. Does anyone still have that, please? It was someone from the University of Chicago, I believe. Thanks in advance. Sincerely, Laura Holt mailto: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Any ideas on how to add a dotted line to a box plot to indicate a specific value?
Basically, i want to create a boxplot from population data, with a solid line for the median, and then a dotted line for a specific persons value. Is there a way to artificially introduce another line or mark of some kind in a box plot? Thanks again for all of yoru help. Hopefully soon I wont have to keep bugging you all. ~Erithid :-p __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] cor vs cor.test
From: ?cor.test Arguments: x, y: numeric vectors of data values. 'x' and 'y' must have the same length. -s. Michael Grant wrote: Using Windows System, R 2.1.0 d is a data frame, 48 rows, 10 columns cor(d) works properly providing all pairwise Pearson correlation coefficients among columns cor.test(d) gives error message Error in cor.test.default(d) : argument y is missing, with no default Why? Thanks, MCG __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] aggregate and stack
Dear All, I have tried to calculate tree mean growth but I think the structure I used below (growthresumo) is not the most elegant, even though it worked. The only problem I had in this first part was that I cannot use 'summary', just 'mean' (sorry but 'R' is pretty new for me). growthresumo - aggregate(growth[,c(16,19,23,27,31,35,39,43,47,52,56,60,64,68,72,76,81,85,89,93,97,101,105,109,113,117,121,125,129,133,137, 141,145,149,153,157,161,165,169,173,177,181,185,189,194,197,201,205,209,213,217,221,225,229,233,237,241)], by=(growth[,c(3,8)]),MEAN,na.rm=TRUE) #after growth is calculated, I want to stack the results in just one colunm. growthvertical - c(growthresumo[,3],...,growthresumo[,50]) # this is very time consuming though Parcel - c(C9,S8...C9,S8) # 50 items date c(DATE1DATE50) growthpermonth - data.frame(Parcel, Date, growthvertical) Thank you very much! Paulo Paulo Brando Inst. de Pesquisa Ambiental da Amazônia (IPAM) Rua Rui Barbosa,136. 68.005.080 Santarém, PA, Brasil. Fone/Fax ++ 55 93 522 5538 www.ipam.org.br [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] global normalization
Hello all, I have question about global normalization. I have data from big experement(two-color cDNA arrays) where used three different layouts but the same set of genes. I used limma package for within array normalization. My question is: how I can do between array normalization (global) if arrays have different layout(different location and different number technical replicates (number of spots of a gene on an array))? Sincerely, Natalia.__ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] global normalization
Please address this question to the appropriate BioConductor or other microarray list (or to the limma package author, as the Posting Guide asks). It is not appropriate for R-Help. -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA The business of the statistician is to catalyze the scientific learning process. - George E. P. Box -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of NATALIA F TCHETCHERINA Sent: Wednesday, May 25, 2005 3:33 PM To: r-help@stat.math.ethz.ch Subject: [R] global normalization Hello all, I have question about global normalization. I have data from big experement(two-color cDNA arrays) where used three different layouts but the same set of genes. I used limma package for within array normalization. My question is: how I can do between array normalization (global) if arrays have different layout(different location and different number technical replicates (number of spots of a gene on an array))? Sincerely, Natalia. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Basic matematical functions with NAs
Conceptually your assumption is wrong: If you have the traps out and don't catch fruit in the trap than you have 0 captures. NA would come for example from a trap not functioning properly at a certain day. To get R to recognize NA you need to define your variables as numeric or factors. Quickest I find to produce the data in a spreadsheet/database and have NAs defined as -9. Than import into R via CSV format and replace -9 with NA. Do read the R documentation on http://www.r-project.org/. Introduction to R is most helpful... Herry --- Alexander Herr - Herry Northern Futures Davies Laboratory, CSIRO PMB, Aitkenvale, QLD 4814 Phone (07) 4753 8510 Fax (07) 4753 8650 Home: http://herry.ausbats.org.au Webadmin ABS: http://ausbats.org.au Sustainable Ecosystems: http://www.cse.csiro.au/ Original messageX Message: 19 Date: Tue, 24 May 2005 10:22:53 -0700 From: Paulo Brando [EMAIL PROTECTED] Subject: [R] Basic matematical functions with NAs To: r-help@stat.math.ethz.ch Message-ID: [EMAIL PROTECTED] Content-Type: text/plain Dear All, I've tried to sum columns -- different species of flowers, fruits plus twigs -- with NAs to get litterfall/trap, and then after use litterfall to calculate production (litterfall (grams)/ hectare/ day. But R 'sees' litterfall/trap as a string. My question: How to use basic mathematical functions to deal with NAs in data management. Example (as you can note I have many missing values -- no fruit fell in the trap. area ponto date pseco psaco pliquido florg1 flor1 florg2 flor2 florg3 flor3 frutog1 fruto1 frutog2 fruto2 frutog3 fruto3 frutog4 fruto4 frutog5 fruto5 frutog6 fruto6 frutog7 fruto7 frutog8 fruto8 twigs A A 1 38233 17.7 1.6 7.1 0.266 1 A AA 1 38233 12.5 8.7 3.8 A AB 1 38233 13.9 1.7 3.2 0.421 3 A B 1 38233 12.1 1.6 1.5 0.248 2 0.435 7 0.16 1 A BORDA 1 38233 A C 1 38233 15.6 1.7 4.9 0.374 2 0.298 3 0.231 1 A F 1 38233 14 1.5 3.5 0.366 45 0.153 1 0.15 1 Paulo Brando Inst. de Pesquisa Ambiental da Amazônia (IPAM) Rua Rui Barbosa,136. 68.005.080 Santarém, PA, Brazil. Fone/Fax ++ 55 93 522 5538 www.ipam.org.br [[alternative HTML version deleted]] - __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Problem with systemfit 0.7-3 and transformed variables
On Wednesday 25 May 2005 18:25, Mikko Pakkanen wrote: We did not notice this shortcoming of systemfit() so far. Unfortunately, I don't have the time in the next few days to look into the code and figure out how to enable transformed variables. I suggest that you either create transformed variables by hand or you modify the systemfit code to enable this and send us the patch. I prefer the second :-) (that's the philosophy of open-source software like R: useRs become developeRs). Luckily, I had some time to check the code. Debugger revealed that the problems are caused by the model.frame function which is used to compile the '$data' data frame. I don't need that data frame so much, so I just substituted model.frame with model.matrix which apparently doesn't cause this error with transformed variables. However, I tuned it a bit further, so that it should still return an identical '$data' data frame, despite the modification. I've only tested this with my example and it appears to be OK. Still, I think this should be considered a quick dirty fix -there are probably better ways to do it. But, I hope it gives the idea. Here's my attempt: [EMAIL PROTECTED] R $ diff -u systemfit.R systemfit-patched.R --- systemfit.R 2004-11-26 11:17:36.0 +0200 +++ systemfit-patched.R 2005-05-25 18:55:55.568944699 +0300 @@ -624,7 +624,11 @@ Terms - terms( eqns[[i]], data = data) m$formula - Terms m - eval(m, parent.frame()) -datai - model.frame(Terms, m) +resp - model.extract(m, response) + ## using model.matrix instead of model.frame, need to get the output variable separately +datai - data.frame(cbind(resp, (model.matrix(Terms, m))[,-1])) + ## I guess there's a better way to extract the name of the output variable? + names(datai)[1] - as.character(terms(eqns[[i]]))[2] if(method==2SLS | method==3SLS) { #datai - cbind( datai, model.frame( instl[[i]] )) # the following lines have to be substituted for the previous @@ -634,7 +638,8 @@ Terms - terms(instl[[i]], data = data) m$formula - Terms m - eval(m, parent.frame()) - datai - cbind( datai, model.frame(Terms, m)) + ## used previously model.frame + datai - cbind( datai, as.data.frame((model.matrix(Terms, m))[,-1])) } if(i==1) { Thank you very much for the patch. I have uploaded a patched version of systemfit (version 0.7-4) to CRAN. It is also available on my website now: http://www.uni-kiel.de/agrarpol/ahenningsen/index-e.html Best wishes, Arne Regards, -Mikko. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Arne Henningsen Department of Agricultural Economics University of Kiel Olshausenstr. 40 D-24098 Kiel (Germany) Tel: +49-431-880 4445 Fax: +49-431-880 1397 [EMAIL PROTECTED] http://www.uni-kiel.de/agrarpol/ahenningsen/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Any ideas on how to add a dotted line to a box plot to indicate a specific value?
Hi BJ wrote: Basically, i want to create a boxplot from population data, with a solid line for the median, and then a dotted line for a specific persons value. Is there a way to artificially introduce another line or mark of some kind in a box plot? Thanks again for all of yoru help. Hopefully soon I wont have to keep bugging you all. ~Erithid :-p Do you mean something like this ... ? with(ToothGrowth, { boxplot(len ~ supp) # boxplots horizontally centred at 1 and 2 # 0.6, 1.4 come from default boxwex of 0.8 # (1 +/- boxwex/2) # 18 I made up lines(c(0.6, 1.4), c(18, 18), lty=dashed, col=red) }) Paul -- Dr Paul Murrell Department of Statistics The University of Auckland Private Bag 92019 Auckland New Zealand 64 9 3737599 x85392 [EMAIL PROTECTED] http://www.stat.auckland.ac.nz/~paul/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Plot: Distance between tick and tick label on y-axis
Hi You might have more luck with par(mgp), for example, ... par(las=1) plot(runif(50), type=l,xaxt=n,yaxt=n,ylab=, bty=l) axis(2, mgp=c(3, .5, 0)) axis(1, mgp=c(3, .3, 0)) Paul Lapointe, Pierre wrote: Hello, I want to reduce the distance between my ticks and their labels. I managed to do it for the x-axis, but the y-axis puzzles me. Here's an example where there is no space between the x-asix ticks and labels. par(las=1) plot(runif(50), type=l,xaxt=n,yaxt=n,ylab=, bty=l) axis(2) axis(1,padj=-1.5) #However, axis(2,padj=-1.5) #does not work I understand from ?axis that padj will take its direction from the par(las). In this case, padj will move labels up and down for both x-axis and y-axis. I want my y-axis labels to be horizontal. I can I reduce the distance between y-axis ticks and labels? Regards, Pierre Lapointe Assistant Market Strategist *** AVIS DE NON-RESPONSABILITE:\ Ce document transmis par courri...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Dr Paul Murrell Department of Statistics The University of Auckland Private Bag 92019 Auckland New Zealand 64 9 3737599 x85392 [EMAIL PROTECTED] http://www.stat.auckland.ac.nz/~paul/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] PAN: Need Help for Multiple Imputation Package
Hello all. I am trying to run PAN, multilevel multiple imputation program, in R to impute missing data in a longitudinal dataset. I could successfully run the multiple imputation when I only imputed one variable. However, when I tried to impute a time-varying covariate as well as a response variable, I received an error message, Error: subscript out of bounds. Can anyone tell if my commands contain any mistakes? First I imported SAS dataset sim which includes a response variable MIY1, a time-varying covariate TCOV1, TIME, GROUP (0 or 1), and ID. 200 participants were included and measurement occurred six times. Approximately 25% of participants dropped out at end. sim - read.xport('c:\\xptds.dat') int - rep(1,1200) y - cbind(sim$MIY1,sim$TCOV1) subj - sim$ID pred - cbind(int, sim$TIME, sim$GROUP) xcol - 1:3 zcol - 1 prior - list(a=2,Binv=4,c=2,Dinv=4) result - pan(y,subj,pred,xcol,zcol,prior,seed=13579,iter=1000) Error: subscript out of bounds By the way, I also received the same error message when I tried to include intercept and time in Zcol, a matrix for random effect specification. I used command zcol - 1:2. Does anybody know if this error is due to sample size/proportion of missing data or due to command mistake? I truly appreciate any feedbacks. Best regards, Eishi __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] precision problem
Dear Omar, It wasn't clear to me from your original question that you wanted to test that *all* the corresponding entries were equal, as opposed to each individual entry. In any event, I don't think that you'll find a similar function for testing inequality, so you can do as you suggest, but of course without abs(). Regards, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Omar Lakkis Sent: Wednesday, May 25, 2005 10:08 AM To: r-help@stat.math.ethz.ch Subject: Re: [R] precision problem all.equal is helpful when I am comparing equality of two matrices. However, when I am comparing two individual number with or is my best bet doing if( abs(x - y) tolerence) or is there a function like all.equal that has the same default tolerence? On 5/25/05, Omar Lakkis [EMAIL PROTECTED] wrote: Thank you all. all.equal is very helpful since I am also interested in finding the mismatched prices. On 5/25/05, John Fox [EMAIL PROTECTED] wrote: Dear Omar, Perhaps I'm missing something, but why not just subtract one matrix from the other and test the difference in relation to the precision that you require for the comparison? E.g., to test near equality, something like, abs(A - B) 1e-13. I hope this helps, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Omar Lakkis Sent: Wednesday, May 25, 2005 9:09 AM To: r-help@stat.math.ethz.ch Subject: [R] precision problem I have prices that I am finding difficult to compare with ==, and , due to precision. For example: the numbers should match, with '==', but they differ in the magnitude of 1e-14 due to bunch of calculations that I run on them. Programming with java, I am used to implementing a function that compares the difference between the numbers to a pre determined precision factor. This could be very slow when I have two matrices of numbers that I could otherwise compare with a simple '==', '' or '' in R. What is teh best solution for this problem? Can I control the precision of ==, and without having to reimplement the operations in a slow way? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Useful tip: Use Google to find R scripts
Hello, Ever wondered how people use a particular function in their programs? Use Google to find actual scripts: filetype:R boxplot will return real R scripts using the boxplot function. Regards, Pierre Lapointe Assistant Market Strategist *** AVIS DE NON-RESPONSABILITE:\ Ce document transmis par courri...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] aggregate and stack
Dear Paulo, On May 25, 2005, at 8:01 PM, Paulo Brando wrote: Dear All, I have tried to calculate tree mean growth but I think the structure I used below (growthresumo) is not the most elegant, even though it worked. The only problem I had in this first part was that I cannot use 'summary', just 'mean' (sorry but 'R' is pretty new for me). In case you didn't notice, help(aggregate) indicates that 'FUN' should be a scalar function, so summary won't work for that reason. growthresumo - aggregate(growth[,c(16,19,23,27,31,35,39,43,47,52,56,60,64,68,72,76,81 ,85,89,93,97,101,105,109,113,117,121,125,129,133,137, 141,145,149,153,157,161,165,169,173,177,181,185,189,194,197,201,205,209 ,213,217,221,225,229,233,237,241)], by=(growth[,c(3,8)]),MEAN,na.rm=TRUE) It's hard to know where 'growth' came from. Is it your own data.frame, or from a package? It's better to provide a reproducible or toy example (as you'll often read here). #after growth is calculated, I want to stack the results in just one colunm. growthvertical - c(growthresumo[,3],...,growthresumo[,50]) # this is very time consuming though This comes to my mind: as.vector(as.matrix(growthresumo[,3:50])) but look up the help on stack() because it's a very powerful tool that is aptly named (and might do everything you want). Parcel - c(C9,S8...C9,S8) # 50 items rep() could help with the above. date c(DATE1DATE50) paste() will help with this. growthpermonth - data.frame(Parcel, Date, growthvertical) Thank you very much! Good luck with R! Stephen __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] estimating p values bootstrap regression
Dear R-users: I have a basic question. I am trying to estimate the gradients of sexual selection on the males of a waterbug species.I performed a bootstrap analysis on a multiple regression using library boot and obtained the estimated coefficents and their respective standard errors. eri-read.table(D:/efi/er2.txt,header=T) grad1-eri grad1$fit-fitted(gradi) grad1$res-resid(gradi) grad.fit-function(data){ mod-lm(data$huevrel~data$slongitud*data$sabdomen*data$saparcap, weights=data$presencia, data=eri) c(coef(mod), summary(mod)$coef[,2]^2)} gmodfun-function(d,i){ d$huevrel-d$fit+d$res[i] grad.fit(d)} grandi-boot(grad1,gmodfun, R=999) grandi I was wondering if there is a way to estimate p-values in order to know wich independent variables are significant. Would be correct to add the bias to the regression coefficients and divide with de SE to obtain a t value and use pt function eg (1 - pt(abs(1.51),211))*2 in order to get the p values? Thanks a lot for your help. Roberto I use R 2.1, Windows XP. Departamento de Biologia Evolutiva Instituto de Ecología, A.C. Km 2.5 Carretera antigua a Coatepec Ap. Postal 63 (excepto mensajería) Xalapa, Veracruz 91000 MEXICO Tel. (52)(228)8421800 ext.3009 - Instituto de Ecología, A. C. http://www.ecologia.edu.mx __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html