[R] scalassoc package
Many new things at http://www.cuddyvalley.org/psychoR/ The scalassoc package, which fits exponential distance association models to indicator matrices, it now at version 1.0.0. It seems to be robust and can analyze large examples easily. It is a major improvement (in speed and robustness) over the distassoc package, which is at the same site. scalassoc does something neat (if you like that sort of thing). It writes the changing configurations to a plotwindow for your default device, showing the iterations as a movie. But, if you have ffmpeg installed in your path, then it also has the option to write the iterations to a quicktime movie file. Currently this creates a lot of intermediate jpeg's (although it cleans up after itself). It may be possible to use ffmpeg to stream them directly into a movie file. Just to give you an idea, the data are in the n x k_j indicator matrices G_j, where g_{ijl}=1 if object i is in category (level) l of variable j. The log- likelihood we maximize is pastedGraphic.pdf Description: Adobe PDF document Variable j has k_j levels, and after we are done we can make a Voronoi diagram (using the deldir package) of the k_j points y_{jl}. Maximizing the likelihood means trying to make sure each of the x_i is in the "correct" Voronoi cell, i.e. the Voronoi cell corresponding with the category of variable j that i was in (i.e. for which g_{ijl}=1). This generalizes multidimensional IRT models, the choice models used for voting data in political science, the Goodman-Haberman-Gilula-Ritov distance association models, the Luce-Shepard choice model, and so on, to multivariate/multicategory data. It is part of the "Gifi Goes Logistic" project. The algorithm is based on majorization, starting with multiple correspondence analysis, and each iteration does one step of a truncated SVD (with a different target in each iteration). The movies show the movement of X from one iteration to the next. === Jan de Leeuw; Distinguished Professor and Chair, UCLA Department of Statistics; Editor: Journal of Multivariate Analysis, Journal of Statistical Software US mail: 8125 Math Sciences Bldg, Box 951554, Los Angeles, CA 90095-1554 phone (310)-825-9550; fax (310)-206-5658; email: [EMAIL PROTECTED] .mac: jdeleeuw ++ aim: deleeuwjan ++ skype: j_deleeuw homepages: http://gifi.stat.ucla.edu ++ http://www.cuddyvalley.org - No matter where you go, there you are. --- Buckaroo Banzai http://gifi.stat.ucla.edu/sounds/nomatter.au - __ 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] Unfound objects in function
A couple of my functions that were working last week seem to have been changed over the weekend and no longer work, but I can't understand why not: it seems that objects defined at the start of the function are not located further on in the function, when this worked fine before. An example follows at the end, using the package spatstat, though the problem seems more general. My only guess is that I've accidentally changed something to do with the environment that functions are searching in to exclude themselves, but if so I've no idea how I did or how I can cancel it! Thanks in advance, Colin library (spatstat) kernelEst <- function (data, max.r = 10, edge = TRUE) { if (!is.ppp (data)) stop ("data must be a point process object") i <- 0.1 diff<- 1 smo <- 1 while (diff > 0) { olddiff <- diff oldsmo <- smo smo <- density.ppp (data, sigma = 2 * i, edge = edge) lambda <- smo[data] Ki <- Kinhom (data[,data$window], lambda = lambda, r = seq (0, max.r, length = 100), correction = "trans") diff<- sum (Ki$theo - Ki$trans) i <- i + 0.1 if (i > max.r) stop ("no suitable kernel found within 0.2 to max.r") } if (abs (olddiff) > abs (diff)) return (smo) else return (oldsmo) } data (simdat) test <- kernelEst (simdat) Error in while (diff > 0) { : missing value where TRUE/FALSE needed > version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.1 year 2005 month12 day 20 svn rev 36812 language R __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] get the parameter estimates from lm model
R-help, Sorry for posting the same question (almost) once again. I'm using a lm model for some data. The code is below: fitData <- glm(pm ~ age + mw , family = binomial(link = logit), maxit = 30, data = inputModel1) "age" is a factor and "mw" is a continous variable. The summary function does not provide the parameter estimates for the first level of the categorical variable (I think the others are relative to the first,isn't it?). How can I get all the parameter estimates in the model? Thank you in advance I run on a Windows Xp machine > version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.1 year 2005 month12 day 20 svn rev 36812 language R __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] get the parameter estimates from lm model - was "glm"
R-help, Sorry for the mistake in my previous message. I have written "lm" but in the function call is "glm". Everything should be "glm". Sorry for the inconvenience. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] get the parameter estimates from lm model
Do you have a look at names(fitData) fitData$coef Best, Matthias > -Ursprüngliche Nachricht- > Von: [EMAIL PROTECTED] [mailto:r-help- > [EMAIL PROTECTED] Im Auftrag von Luis Ridao Cruz > Gesendet: Dienstag, 18. April 2006 10:23 > An: r-help@stat.math.ethz.ch > Betreff: [R] get the parameter estimates from lm model > > R-help, > > Sorry for posting the same question (almost) once again. > > I'm using a lm model for some data. > The code is below: > > fitData <- glm(pm ~ age + mw , family = binomial(link = logit), maxit = > 30, data = inputModel1) > > "age" is a factor and "mw" is a continous variable. > > The summary function does not provide the parameter estimates for > the first level of the categorical variable (I think the others are > relative to the first,isn't it?). > > How can I get all the parameter estimates in the model? > > > Thank you in advance > > > > I run on a Windows Xp machine > > > version > _ > platform i386-pc-mingw32 > arch i386 > os mingw32 > system i386, mingw32 > status > major2 > minor2.1 > year 2005 > month12 > day 20 > svn rev 36812 > language R > > __ > 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] get the parameter estimates from lm model
Look at: ?contrasts ?contr.treatment 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/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: "Luis Ridao Cruz" <[EMAIL PROTECTED]> To: Sent: Tuesday, April 18, 2006 10:22 AM Subject: [R] get the parameter estimates from lm model > R-help, > > Sorry for posting the same question (almost) once again. > > I'm using a lm model for some data. > The code is below: > > fitData <- glm(pm ~ age + mw , family = binomial(link = logit), > maxit = > 30, data = inputModel1) > > "age" is a factor and "mw" is a continous variable. > > The summary function does not provide the parameter estimates for > the first level of the categorical variable (I think the others are > relative to the first,isn't it?). > > How can I get all the parameter estimates in the model? > > > Thank you in advance > > > > I run on a Windows Xp machine > >> version > _ > platform i386-pc-mingw32 > arch i386 > os mingw32 > system i386, mingw32 > status > major2 > minor2.1 > year 2005 > month12 > day 20 > svn rev 36812 > language R > > __ > 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 > Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ 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] Unfound objects in function
It does not say `unfound' it says `missing value'. See ?NA. Note it is the value and not the object that is said to be missing. Looks like on your data that one or more of the values you are summing is NA. On Tue, 18 Apr 2006, Colin Beale wrote: > A couple of my functions that were working last week seem to have been > changed over the weekend and no longer work, but I can't understand why > not: it seems that objects defined at the start of the function are not > located further on in the function, when this worked fine before. An > example follows at the end, using the package spatstat, though the > problem seems more general. My only guess is that I've accidentally > changed something to do with the environment that functions are > searching in to exclude themselves, but if so I've no idea how I did or > how I can cancel it! > > Thanks in advance, > > Colin > > > library (spatstat) > > kernelEst <- function (data, max.r = 10, edge = TRUE) { > > if (!is.ppp (data)) stop ("data must be a point process object") > > i <- 0.1 > diff<- 1 > smo <- 1 > > while (diff > 0) { >olddiff <- diff >oldsmo <- smo >smo <- density.ppp (data, sigma = 2 * i, edge = edge) >lambda <- smo[data] >Ki <- Kinhom (data[,data$window], lambda = lambda, r = seq (0, > max.r, length = 100), correction = "trans") >diff<- sum (Ki$theo - Ki$trans) >i <- i + 0.1 >if (i > max.r) stop ("no suitable kernel found within 0.2 to > max.r") > } > > if (abs (olddiff) > abs (diff)) return (smo) else return (oldsmo) > } > > data (simdat) > test <- kernelEst (simdat) > Error in while (diff > 0) { : missing value where TRUE/FALSE needed > > > >> version > _ > platform i386-pc-mingw32 > arch i386 > os mingw32 > system i386, mingw32 > status > major2 > minor2.1 > year 2005 > month12 > day 20 > svn rev 36812 > language R > > __ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Extra \input command in Sweave created figures.
Dear list, I am using Sweave to generate figures and tables for my thesis. Now, I have run into the problem that all of a sudden, Sweave produces not only an \includegraphics command, but a \include command as well. This breakes LaTeX processing. This is an example: The code: \begin{figure}[htbp] \centering <>= sel <- mannerDF$target %in% c("skal","kal","sal","gal") & mannerDF$onsetLength == 1 print( dotplot( prop.table(table(mannerDF$agem[sel],mannerDF$target[sel],mannerDF$manner[sel]),1) ,auto.key=list( space="right" ) ) ) @ \caption[Relative frequence of occurence for each manner of articulation in simple productions of Skal,kal,gal and sal]{Relative frequence of occurence for each manner of articulation in simple productions of the skal,kal,gal and sal target words.} \label{fig:manner-skal-simple} \end{figure} Produces this LaTeX output when run through Sweave: \begin{figure}[htbp] \centering \input{manner-mannerSimpleSkal} \includegraphics{manner-mannerSimpleSkal} \caption[Relative frequence of occurence for each manner of articulation in simple productions of Skal,kal,gal and sal]{Relative frequence of occurence for each manner of articulation in simple productions of the skal,kal,gal and sal target words.} \label{fig:manner-skal-simple} \end{figure} What is causing the extra \input? I had this problem once before, but was enable then to fix the problem by randomly adding and removing blank space. Now, this problem seem to have occured in 18 figures, which means I would like to find the acctual problem this time rather than trying to remove the symptoms. Thankful for all the help I could get. /Fredrik __ 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] adding lines to scatterplot3d
Hi to all, I usually use R for simple 2D graphs so now I cannot solve the following problem (which I suppose very simple too). I'm using scatterplot3d. I want to add dotted lines parallel to the 3 axes, in order to create a sort of cage in the 3D box (so to represent subdomains in my dataspace). I saw that points3d can be used to draw points or lines to the existing plot but actually I'm able only to draw points (I grabbed an example from the web) Could anyone post a line of code with an example? Thanks a lot -a- Andrea Valle DAMS - Facoltà di Scienze della Formazione Università degli Studi di Torino http://www.semiotiche.it/andrea [EMAIL PROTECTED] [[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
Re: [R] using betareg: problems with anova and predict
On Mon, 17 Apr 2006, Paul Johnson wrote: > Dear R-helpers: > > We have had fun using betareg to fit models with proportions as > dependent variables. > > However, in the analysis of these models we found some wrinkles and > don't know where is the best place to start looking for a fix. > > The problems we see (so far) are that > > 1. predict ignores newdata The reason for this is that predict.betareg() does not take a newdate argument, you need to use the terms argument instead. > 2. anova does not work ...because there is no anova() method for betareg objects. I think it is owrth addressing both issues in betareg, I'll write to you and Alexandre (the package maintainer) off-list. Z > Here is the small working example: > > > x <- c(1, 3, 1, 4, 5, 3, 5, 2, 5, 2) > y <- c(.3, .4, .4, .5, , .7, .4, .3, .4, .3, .5) > x2 <- c( 4, 2, 1, 5, 1, 2, 4, 2, 1, 3) > library(betareg) > mybreg <- betareg(y~x) > summary(mybreg) > > predict(mybreg, newdata=data.frame(x=c(2,2))) > > mybreg2 <- betareg(y~x+x2) > summary(mybreg) > anova(mybreg, mybreg2) > > --- > Example output: > > > predict(mybreg, newdata=data.frame(x=c(2,2))) > [1] 0.3903155 0.4207632 0.3903155 0.4362319 0.4518258 0.4207632 0.4518258 > [8] 0.4054484 0.4518258 0.4054484 > ... > > anova(mybreg, mybreg2) > Error in as.double.default(lapply(objects, df.residual)) : > (list) object cannot be coerced to 'double' > > I have been digging in this a little bit and notice betareg does not > return the log likelihood at the maximum likelihood estimates, but I > am able to hack the file /usr/lib/R/library/betareg/R/betareg and save > the "value" of the optim function and print that out, so at least I > can do a likelihood-ratio test by hand in place of the anova. But I'm > stumped on why the predict ignores newdata. > > I'm using R-2.1.1 on Fedora Core Linux 5 with the betareg package from CRAN. > > In case you are interested in betareg, we have found this article to > be extremely readable and helpful: FERRARI, S.L.P., CRIBARI-NETO, F. > (2004). Beta regression for modeling rates and proportions. Journal > of Applied Statistics, v. 31, n. 7, p. 799-815. > > > -- > Paul E. Johnson > Professor, Political Science > 1541 Lilac Lane, Room 504 > University of Kansas > > __ > 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] Subset dataframe based on condition
Thanx Steve and Tony for your help. Sachin Tony Plate <[EMAIL PROTECTED]> wrote: Works OK for me: > x <- data.frame(a=10^(-2:7), b=10^(10:1)) > subset(x, a > 1) a b 4 1e+01 1e+07 5 1e+02 1e+06 6 1e+03 1e+05 7 1e+04 1e+04 8 1e+05 1e+03 9 1e+06 1e+02 10 1e+07 1e+01 > subset(x, a > 1 & b < a) a b 8 1e+05 1000 9 1e+06 100 10 1e+07 10 > Do you get all "numeric" for the following? > sapply(x, class) a b "numeric" "numeric" > If not, then your data frame is probably encoding the information in some way that you don't want (though if it was as factors, I would have expected a warning from the comparison operator). You might get more help by distilling your problem to a simple example that can be tried out by others. -- Tony Plate Sachin J wrote: > Hi, > > I am trying to extract subset of data from my original data frame > based on some condition. For example : (mydf -original data frame, submydf > - subset dada frame) > > >submydf = subset(mydf, a > 1 & b <= a), > > here column a contains values ranging from 0.01 to 10. I want to > extract only those matching condition 1 i.e a > . But when i execute > this command it is not giving me appropriate result. The subset df - > submydf contains rows with 0.01 also. Please help me to resolve this > problem. > > Thanks in advance. > > Sachin > > > - > > [[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 > - [[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] Extra \input command in Sweave created figures.
Fredrik Karlsson gmail.com> writes: > I am using Sweave to generate figures and tables for my thesis. Now, I > have run into the problem that all of a sudden, Sweave produces not > only an \includegraphics command, but a > \include command as well. This breakes LaTeX processing. > Check if you have set split=TRUE in the Sweave options. Dieter __ 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] adding lines to scatterplot3d
andrea valle di.unito.it> writes: > > I'm using scatterplot3d. I want to add dotted lines parallel to the 3 > axes, in order to create a sort of cage in the 3D box (so to represent > subdomains in my dataspace). > I saw that points3d can be used to draw points or lines to the existing > plot but actually I'm able only to draw points (I grabbed an example > from the web) See example 5 in the documentation, and use type="l" to produce connecting lines. Dieter __ 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] adding lines to scatterplot3d
Thanks a lot, I having it working. Is it possible to have dashed or dotted lines? I wasn't able to find a fitting type. Best -a- On 18 Apr 2006, at 14:42, Dieter Menne wrote: > andrea valle di.unito.it> writes: > >> >> I'm using scatterplot3d. I want to add dotted lines parallel to the 3 >> axes, in order to create a sort of cage in the 3D box (so to represent >> subdomains in my dataspace). >> I saw that points3d can be used to draw points or lines to the >> existing >> plot but actually I'm able only to draw points (I grabbed an example >> from the web) > > See example 5 in the documentation, and use type="l" to produce > connecting lines. > > Dieter > > __ > 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 > > Andrea Valle DAMS - Facoltà di Scienze della Formazione Università degli Studi di Torino http://www.semiotiche.it/andrea [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] adding lines to scatterplot3d
andrea valle wrote: > Thanks a lot, I having it working. > > Is it possible to have dashed or dotted lines? > I wasn't able to find a fitting type. All as in standard plot functions: lty="dotted" or lty="dashed" should do the trick. Uwe Ligges > Best > > -a- > > On 18 Apr 2006, at 14:42, Dieter Menne wrote: > > >>andrea valle di.unito.it> writes: >> >> >>>I'm using scatterplot3d. I want to add dotted lines parallel to the 3 >>>axes, in order to create a sort of cage in the 3D box (so to represent >>>subdomains in my dataspace). >>>I saw that points3d can be used to draw points or lines to the >>>existing >>>plot but actually I'm able only to draw points (I grabbed an example >>>from the web) >> >>See example 5 in the documentation, and use type="l" to produce >>connecting lines. >> >>Dieter >> >>__ >>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 >> >> > > Andrea Valle > DAMS - Facoltà di Scienze della Formazione > Università degli Studi di Torino > http://www.semiotiche.it/andrea > [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
Re: [R] adding lines to scatterplot3d
Ah! It does. Thanks a lot. (I use R for batch processing, and I when I discover the right setup I don't change it for months... so I always forget basics...sorry) Best -a- On 18 Apr 2006, at 15:18, Uwe Ligges wrote: > andrea valle wrote: > >> Thanks a lot, I having it working. >> Is it possible to have dashed or dotted lines? >> I wasn't able to find a fitting type. > > > All as in standard plot functions: > > lty="dotted" or lty="dashed" should do the trick. > > Uwe Ligges > > > >> Best >> -a- >> On 18 Apr 2006, at 14:42, Dieter Menne wrote: >>> andrea valle di.unito.it> writes: >>> >>> I'm using scatterplot3d. I want to add dotted lines parallel to the 3 axes, in order to create a sort of cage in the 3D box (so to represent subdomains in my dataspace). I saw that points3d can be used to draw points or lines to the existing plot but actually I'm able only to draw points (I grabbed an example from the web) >>> >>> See example 5 in the documentation, and use type="l" to produce >>> connecting lines. >>> >>> Dieter >>> >>> __ >>> 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 >>> >>> >> Andrea Valle >> DAMS - Facoltà di Scienze della Formazione >> Università degli Studi di Torino >> http://www.semiotiche.it/andrea >> [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 > > Andrea Valle DAMS - Facoltà di Scienze della Formazione Università degli Studi di Torino http://www.semiotiche.it/andrea [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] how to change legend size in a figure
Dear All, I am producing a figure with many curves on it. How do I make the legends for all those curves smaller so that it can fit the figure itself? The commands I used for ploting are: plot(x1,y1,col=1,lty=1) lines(x2,y2,col=2,lty=2) ... legend(0.3,0.4,c("name1","name2",...),col=1:20,lty=1:20) Any tips for making the legend fit the figure will very welcome! Thanks! Best, Baoqiang Cao __ 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] lambda, uncertainty coefficient (& Somers D)
Dear colleagues in R, Has anybody implemented the 1) (Goodman & Kruskal) lambda or the 2) (Thiel's) uncertainty coefficient statistics (in the asymmetric and symmetric forms), or is anyone aware that they might reside in some package? A search in the R archives does indicate that they are (somehow) part of the CoCo package, but I would rather not start learning how to transform my data into CoCo-format in order to access these functions, regardless of whether the CoCo versions are actually intended for calculating the actual statistic or for some other package internal purposes, as may sometimes be the case. Furthermore, it appears to me that the 'somers2' function in the Hmisc package applies Somers' D only to 2x2 and not larger tables. Am I mistaken, or does there exist somewhere else an implementation of the Somers' D statistic for the more general RxC tables? This was queried in 1999, but no response seemed then to be forthcoming. Thanks and regards, -Antti Arppe -- == Antti Arppe - Master of Science (Engineering) Researcher & doctoral student (Linguistics) E-mail: [EMAIL PROTECTED] WWW: http://www.ling.helsinki.fi/~aarppe __ 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 change legend size in a figure
cex=0.8 (or whatever you like) as in legend(0.3,0.4,c("name1","name2",...),col=1:20,lty=1:20, cex=0.8) Charles Annis, P.E. [EMAIL PROTECTED] phone: 561-352-9699 eFax: 614-455-3265 http://www.StatisticalEngineering.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Baoqiang Cao Sent: Tuesday, April 18, 2006 9:38 AM To: r-help@stat.math.ethz.ch Subject: [R] how to change legend size in a figure Dear All, I am producing a figure with many curves on it. How do I make the legends for all those curves smaller so that it can fit the figure itself? The commands I used for ploting are: plot(x1,y1,col=1,lty=1) lines(x2,y2,col=2,lty=2) ... legend(0.3,0.4,c("name1","name2",...),col=1:20,lty=1:20) Any tips for making the legend fit the figure will very welcome! Thanks! Best, Baoqiang Cao __ 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] Nonlinear Regression model: Diagnostics
Hi, I am trying to run the following nonlinear regression model. > nreg <- nls(y ~ exp(-b*x), data = mydf, start = list(b = 0), alg = "default", trace = TRUE) OUTPUT: 24619327 : 0 24593178 : 0.0001166910 24555219 : 0.0005019005 24521810 : 0.001341571 24500774 : 0.002705402 24490713 : 0.004401078 24486658 : 0.00607728 24485115 : 0.007484372 24484526 : 0.008552635 24484298 : 0.009314779 24484208 : 0.009837009 24484172 : 0.01018542 24484158 : 0.01041381 24484152 : 0.01056181 24484150 : 0.01065700 24484149 : 0.01071794 24484148 : 0.01075683 24484148 : 0.01078161 24484148 : 0.01079736 24484148 : 0.01080738 24484148 : 0.01081374 Nonlinear regression model model: y ~ exp(-b * x) data: mydf b 0.01081374 residual sum-of-squares: 24484148 My question is how do I interpret the results of this model. > profile(nreg) 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : Error in prof$getProfile() : number of iterations exceeded maximum of 50 I am unable to understand the error cause. Any pointers would be of great help. Regards, Sachin - [[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] how to change legend size in a figure
Use a smaller cex within legend (e.g. cex=0.75) to adjust the size of the printed text. This controls the size of the legend. Also, you can play with xjust and yjust for finer control of how the legend box is justified at the specified position. -Christos -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Baoqiang Cao Sent: Tuesday, April 18, 2006 9:38 AM To: r-help@stat.math.ethz.ch Subject: [R] how to change legend size in a figure Dear All, I am producing a figure with many curves on it. How do I make the legends for all those curves smaller so that it can fit the figure itself? The commands I used for ploting are: plot(x1,y1,col=1,lty=1) lines(x2,y2,col=2,lty=2) ... legend(0.3,0.4,c("name1","name2",...),col=1:20,lty=1:20) Any tips for making the legend fit the figure will very welcome! Thanks! Best, Baoqiang Cao __ 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] how to change legend size in a figure
probably the 'cex' and 'ncol' arguments of legend() will be useful here. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: "Baoqiang Cao" <[EMAIL PROTECTED]> To: Sent: Tuesday, April 18, 2006 3:38 PM Subject: [R] how to change legend size in a figure > Dear All, > > I am producing a figure with many curves on it. How do I make the > legends for all those curves smaller so that it can fit the figure > itself? The commands I used for ploting are: > > plot(x1,y1,col=1,lty=1) > lines(x2,y2,col=2,lty=2) > ... > legend(0.3,0.4,c("name1","name2",...),col=1:20,lty=1:20) > > Any tips for making the legend fit the figure will very welcome! > Thanks! > > Best, > Baoqiang Cao > > __ > 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 > Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ 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] Unfound objects in function
Colin, Did you use correction = "trans" in Kinhom in the past? I think Ki$trans will contain a number of NAs. Try diff<- sum (Ki$theo - Ki$trans, na.rm = TRUE) Peter Ehlers Colin Beale wrote: > A couple of my functions that were working last week seem to have been > changed over the weekend and no longer work, but I can't understand why > not: it seems that objects defined at the start of the function are not > located further on in the function, when this worked fine before. An > example follows at the end, using the package spatstat, though the > problem seems more general. My only guess is that I've accidentally > changed something to do with the environment that functions are > searching in to exclude themselves, but if so I've no idea how I did or > how I can cancel it! > > Thanks in advance, > > Colin > > > library (spatstat) > > kernelEst <- function (data, max.r = 10, edge = TRUE) { > > if (!is.ppp (data)) stop ("data must be a point process object") > > i <- 0.1 > diff<- 1 > smo <- 1 > > while (diff > 0) { > olddiff <- diff > oldsmo <- smo > smo <- density.ppp (data, sigma = 2 * i, edge = edge) > lambda <- smo[data] > Ki <- Kinhom (data[,data$window], lambda = lambda, r = seq (0, > max.r, length = 100), correction = "trans") > diff<- sum (Ki$theo - Ki$trans) > i <- i + 0.1 > if (i > max.r) stop ("no suitable kernel found within 0.2 to > max.r") > } > > if (abs (olddiff) > abs (diff)) return (smo) else return (oldsmo) > } > > data (simdat) > test <- kernelEst (simdat) > Error in while (diff > 0) { : missing value where TRUE/FALSE needed > > > > >>version > > _ > platform i386-pc-mingw32 > arch i386 > os mingw32 > system i386, mingw32 > status > major2 > minor2.1 > year 2005 > month12 > day 20 > svn rev 36812 > language R > > __ > 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] MLE
cs.uct.ac.za> writes: > > > Hi , > > I want to compute the MLE for a simple sample of data, say > 45,26,98,65,25,36,42,62,28,36,15,48,45, of which I obviously have the mean > and the sd. Is there a way of calling the log normal and already > diffrentiated formula other than entering the whole formula. > > Victor > You haven't given us enough information/described your problem clearly enough. The maximum likelihood estimate of _what_? fitdistr() in the MASS package might help. If you want (log)likelihoods dlnorm (for log normal), dnorm might help. Ben Bolker __ 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] lambda, uncertainty coefficient (& Somers D)
Antti Arppe wrote: > Dear colleagues in R, > > Has anybody implemented the > > 1) (Goodman & Kruskal) lambda > > or the > > 2) (Thiel's) uncertainty coefficient > > statistics (in the asymmetric and symmetric forms), or is anyone aware > that they might reside in some package? A search in the R archives > does indicate that they are (somehow) part of the CoCo package, but I > would rather not start learning how to transform my data into > CoCo-format in order to access these functions, regardless of whether > the CoCo versions are actually intended for calculating the actual > statistic or for some other package internal purposes, as may > sometimes be the case. > > Furthermore, it appears to me that the 'somers2' function in the Hmisc > package applies Somers' D only to 2x2 and not larger tables. Am I > mistaken, or does there exist somewhere else an implementation of the > Somers' D statistic for the more general RxC tables? This was queried > in 1999, but no response seemed then to be forthcoming. Somers' D is for a 2-category response variable. X may have any number of categories as long as they are ordered. Frank Harrell > > Thanks and regards, > > -Antti Arppe -- 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] how funciton "expression" produces subscript
Dear R-users, does anyone know how funciton "expression" produces subscript? Best,Jing __ 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 funciton "expression" produces subscript
See ?plotmath for the answer to your question and many more regarding math expression drawing. It might be worth printing this help page for future reference. -Christos -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jing Yang Sent: Tuesday, April 18, 2006 10:24 AM To: r-help Subject: [R] how funciton "expression" produces subscript Dear R-users, does anyone know how funciton "expression" produces subscript? Best,Jing __ 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] what happen?
Hi! I want to choose part of the data. But it has a question. > attach(PIKO) > PIKO[1:5,] code species dbh x y tag status branch 92 10142PIKO 38.9 6.05 12.81 10165 alive 0 109 10213PIKO 41.0 6.71 26.21 10202 alive 0 135 10222PIKO 48.3 18.98 21.28 10214 alive 0 146 10223PIKO 47.0 19.45 22.83 10224 alive 0 147 10223PIKO 20.0 19.74 28.65 10225 alive 0 > n=PIKO[dbh>=25,] > n[1:5,] code species dbh x y tag status branch 92 10142PIKO 38.9 6.05 12.81 10165 alive 0 109 10213PIKO 41.0 6.71 26.21 10202 alive 0 135 10222PIKO 48.3 18.98 21.28 10214 alive 0 146 10223PIKO 47.0 19.45 22.83 10224 alive 0 152 10223PIKO 33.5 16.94 27.25 10230 alive 0 *> n=PIKO[status="snag",] Error in "[.data.frame"(PIKO, status = "snag", ) : unused argument(s) (status ...) * [[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] Question about PLS regression
Hello useRs, I am new user to R and also statistics. Why predicted results in this example are different? Is the order of variables in X matrix important? library (pls) set.seed (1) Y1 <- c(1,2,3,4,5,6,7,8,9,10) Y2 <- c(0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.0) X1 <- rnorm(10,sd=0.2) X2 <- rnorm(10,sd=1) X3 <- rnorm(10,sd=0.1) X4 <- rnorm(10,sd=0.1) X5 <- rnorm(10,sd=0.1) KAL <- data.frame(num=c(1:10)) KAL$Y <- as.matrix(cbind (Y1,Y2)) KAL$X <- as.matrix(cbind (X1,X2,X3,X4,X5)) KAL2 <- data.frame(num=c(1:10)) KAL2$Y <- as.matrix(cbind (Y1,Y2)) KAL2$X <- as.matrix(cbind (X5,X4,X3,X2,X1)) PLS <- plsr (Y~X,data=KAL, 4,validation = "CV") PLS2 <- plsr (Y~X,data=KAL2,4, validation = "CV") X1v <- rnorm(10,sd=0.1) X2v <- rnorm(10,sd=1) X3v <- rnorm(10,sd=0.1) X4v <- rnorm(10,sd=0.1) X5v <- rnorm(10,sd=0.1) VAL <- data.frame(num=c(1:10)) VAL$X <- as.matrix(cbind(X1v,X2v,X3v,X4v,X5v)) predict (PLS,VAL,4) predict (PLS2,VAL,4) Thank You, Andris Jankevics __ 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] what happen?
Look at: ?Comparison; you need: PIKO[status == "snag", ] Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: "zhang jian" <[EMAIL PROTECTED]> To: "r-help" Sent: Tuesday, April 18, 2006 4:45 PM Subject: [R] what happen? > Hi! > I want to choose part of the data. But it has a question. > >> attach(PIKO) >> PIKO[1:5,] > code species dbh x y tag status branch > 92 10142PIKO 38.9 6.05 12.81 10165 alive 0 > 109 10213PIKO 41.0 6.71 26.21 10202 alive 0 > 135 10222PIKO 48.3 18.98 21.28 10214 alive 0 > 146 10223PIKO 47.0 19.45 22.83 10224 alive 0 > 147 10223PIKO 20.0 19.74 28.65 10225 alive 0 >> n=PIKO[dbh>=25,] >> n[1:5,] > code species dbh x y tag status branch > 92 10142PIKO 38.9 6.05 12.81 10165 alive 0 > 109 10213PIKO 41.0 6.71 26.21 10202 alive 0 > 135 10222PIKO 48.3 18.98 21.28 10214 alive 0 > 146 10223PIKO 47.0 19.45 22.83 10224 alive 0 > 152 10223PIKO 33.5 16.94 27.25 10230 alive 0 > *> n=PIKO[status="snag",] > Error in "[.data.frame"(PIKO, status = "snag", ) : >unused argument(s) (status ...) > * > > [[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 > Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ 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 about PLS regression
I am sorry. It was my fault. My example is wrong. I need also rearrange a validation data set too. But I have a sligthy different results with my real data. Where can the problem be? Andris Jankevics On Otrdiena, 18. Aprīlis 2006 17:55, Andris Jankevics wrote: > Hello useRs, > > I am new user to R and also statistics. Why predicted results in this > example are different? Is the order of variables in X matrix important? > > library (pls) > set.seed (1) > Y1 <- c(1,2,3,4,5,6,7,8,9,10) > Y2 <- c(0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.0) > X1 <- rnorm(10,sd=0.2) > X2 <- rnorm(10,sd=1) > X3 <- rnorm(10,sd=0.1) > X4 <- rnorm(10,sd=0.1) > X5 <- rnorm(10,sd=0.1) > > KAL <- data.frame(num=c(1:10)) > KAL$Y <- as.matrix(cbind (Y1,Y2)) > KAL$X <- as.matrix(cbind (X1,X2,X3,X4,X5)) > KAL2 <- data.frame(num=c(1:10)) > KAL2$Y <- as.matrix(cbind (Y1,Y2)) > KAL2$X <- as.matrix(cbind (X5,X4,X3,X2,X1)) > > PLS <- plsr (Y~X,data=KAL, 4,validation = "CV") > PLS2 <- plsr (Y~X,data=KAL2,4, validation = "CV") > > X1v <- rnorm(10,sd=0.1) > X2v <- rnorm(10,sd=1) > X3v <- rnorm(10,sd=0.1) > X4v <- rnorm(10,sd=0.1) > X5v <- rnorm(10,sd=0.1) > VAL <- data.frame(num=c(1:10)) > VAL$X <- as.matrix(cbind(X1v,X2v,X3v,X4v,X5v)) > > predict (PLS,VAL,4) > predict (PLS2,VAL,4) > > Thank You, > > Andris Jankevics > > __ > 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] 3D pie
Hi all, Is there a way to draw 3D pie with R (like excel does)? I know how to do it in 2D, just by using pie(something)... I know it isn't the best way to represent data, but people are sometimes more interested by the look and feel than by the accuracy of the results... If there is no way, have you another suggestion ? (i already use dotchart instead of pie) Thks to all of you. COMTE Guillaume [[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] lambda, uncertainty coefficient (& Somers D)
On Tue, 2006-04-18 at 16:40 +0300, Antti Arppe wrote: > Dear colleagues in R, > > Has anybody implemented the > > 1) (Goodman & Kruskal) lambda > > or the > > 2) (Thiel's) uncertainty coefficient > > statistics (in the asymmetric and symmetric forms), or is anyone aware > that they might reside in some package? A search in the R archives > does indicate that they are (somehow) part of the CoCo package, but I > would rather not start learning how to transform my data into > CoCo-format in order to access these functions, regardless of whether > the CoCo versions are actually intended for calculating the actual > statistic or for some other package internal purposes, as may > sometimes be the case. > > Furthermore, it appears to me that the 'somers2' function in the Hmisc > package applies Somers' D only to 2x2 and not larger tables. Am I > mistaken, or does there exist somewhere else an implementation of the > Somers' D statistic for the more general RxC tables? This was queried > in 1999, but no response seemed then to be forthcoming. > > Thanks and regards, > > -Antti Arppe Antti, Under a separate e-mail to you, I have sent a text file containing R code for the above and other related association measures. If anyone else is interested in these, please let me know and I will be more than happy to forward them on. I did not want to consume a lot of bandwidth here. HTH, Marc Schwartz __ 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 and ViM
Dear all, I'm starting to learn R, but I'm already programing for a while, using ViM as editor. Therefore I'd like to be able to use R together with ViM. My question now is, whether there are already people out there knowing how to do this in a similar easy way as with Emacs, and if those would be willing to share this knowledge. I did already research on the web on this topic, but i couldn't find satisfying answers, except one good looking approach on the ViM-website with a perl-script called funnel.pl, which I couldn't make running on my mac OSX 10.3.9 so far. I'd be grateful for help. Thanks, Michael Graber __ 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 and ViM
Michael Graber wrote: > Dear all, > > I'm starting to learn R, but I'm already programing for a while, using > ViM as editor. Therefore I'd like to be able to use R together with > ViM. > > My question now is, whether there are already people out there knowing > how to do this in a similar easy way as with Emacs, and if those would > be willing to share this knowledge. > I did already research on the web on this topic, but i couldn't find > satisfying answers, except one good looking approach on the ViM-website > with a perl-script called funnel.pl, which I couldn't make running on > my mac OSX 10.3.9 so far. > > I'd be grateful for help. > > Thanks, > Michael Graber > > __ > 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 Try RSiteSearch('vim editor') at the R prompt. I got 28 hits. 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
[R] store levels in a string
Dear R-users, i need to store in a variable a string made from levels of a factor e.g. a<-("a","a","b","b") af<-factor(a) mylevels<- ...a string with all the levels(af) Thanks -- DM __ 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] lambda, uncertainty coefficient (& Somers D)
On Tue, 2006-04-18 at 10:30 -0500, Marc Schwartz (via MN) wrote: > On Tue, 2006-04-18 at 16:40 +0300, Antti Arppe wrote: > > Dear colleagues in R, > > > > Has anybody implemented the > > > > 1) (Goodman & Kruskal) lambda > > > > or the > > > > 2) (Thiel's) uncertainty coefficient > > > > statistics (in the asymmetric and symmetric forms), or is anyone aware > > that they might reside in some package? A search in the R archives > > does indicate that they are (somehow) part of the CoCo package, but I > > would rather not start learning how to transform my data into > > CoCo-format in order to access these functions, regardless of whether > > the CoCo versions are actually intended for calculating the actual > > statistic or for some other package internal purposes, as may > > sometimes be the case. > > > > Furthermore, it appears to me that the 'somers2' function in the Hmisc > > package applies Somers' D only to 2x2 and not larger tables. Am I > > mistaken, or does there exist somewhere else an implementation of the > > Somers' D statistic for the more general RxC tables? This was queried > > in 1999, but no response seemed then to be forthcoming. > > > > Thanks and regards, > > > > -Antti Arppe > > > Antti, > > Under a separate e-mail to you, I have sent a text file containing R > code for the above and other related association measures. > > If anyone else is interested in these, please let me know and I will be > more than happy to forward them on. I did not want to consume a lot of > bandwidth here. > > HTH, > > Marc Schwartz [Replying to r-help only] Antti, Each of my e-mails directly to you, including the code file, using either: [EMAIL PROTECTED] (located on your web page) or [EMAIL PROTECTED] (used here) have bounced, with access denied/user unknown errors. Can you please confirm a correct e-mail address? Thanks, Marc __ 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] store levels in a string
On Tue, 2006-04-18 at 17:52 +0200, Daniele Medri wrote: > Dear R-users, > > i need to store in a variable a string made from levels of a factor > > e.g. > > a<-("a","a","b","b") The above should be: a <- c("a","a","b","b") ^ > af<-factor(a) > > mylevels<- ...a string with all the levels(af) > > Thanks > -- > DM > mylevels <- levels(af) > mylevels [1] "a" "b" See ?levels, which is listed in the "See Also" for ?factor. HTH, Marc Schwartz __ 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] store levels in a string
Il giorno mar, 18/04/2006 alle 11.00 -0500, Marc Schwartz (via MN): > > mylevels<- ...a string with all the levels(af) > > mylevels <- levels(af) > > > mylevels > [1] "a" "b" I don't need to store these two levels, but a string with the values "ab". Thanks Cheers -- DM __ 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 and ViM
On 18-Apr-06 Sundar Dorai-Raj wrote: > Michael Graber wrote: >> I'm starting to learn R, but I'm already programing for >> a while, using ViM as editor. Therefore I'd like to be >> able to use R together with ViM. >> [...] > > Try RSiteSearch('vim editor') at the R prompt. I got 28 hits. > > HTH, > > --sundar And, out of interest (being a vim addict), I did that, getting > RSiteSearch('vim editor') A search query has been submitted to http://search.r-project.org The results page should open in your browser shortly which it duly did. However -- and this is what's interesting -- I entered the command in R running on my machine B, at which I'm sitting. This machine B is independently displaying the Firefox browser running on Machine C (via X Windows; Linux on both machines). And that's the browser where the search results appeared. So R running on B can hijack Firefox on C. How does it know? Hmmm ... Ted. E-Mail: (Ted Harding) <[EMAIL PROTECTED]> Fax-to-email: +44 (0)870 094 0861 Date: 18-Apr-06 Time: 17:24:22 -- XFMail -- __ 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] store levels in a string
On Tue, 2006-04-18 at 18:07 +0200, Daniele Medri wrote: > Il giorno mar, 18/04/2006 alle 11.00 -0500, Marc Schwartz (via MN): > > > mylevels<- ...a string with all the levels(af) > > > mylevels <- levels(af) > > > > > mylevels > > [1] "a" "b" > > I don't need to store these two levels, but a string with the values > "ab". > > Thanks > > Cheers > -- > DM > paste(levels(af), collapse = "") [1] "ab" See ?paste. HTH, Marc Schwartz __ 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] second try; writing user-defined GLM link function
Is this something that is fairly widely useful? If so, it is something that we could add to the stats package. (We ought in any case to make it easier to extend glm families, and this could be an example. If so, I'd like a reference for its use.) There is a good reason why logit is implmented as > make.link("logit") $linkfun function (mu) .Call("logit_link", mu, PACKAGE = "stats") $linkinv function (eta) .Call("logit_linkinv", eta, PACKAGE = "stats") $mu.eta function (eta) .Call("logit_mu_eta", eta, PACKAGE = "stats") $valideta function (eta) TRUE which is that these functions protect against underflow/overflow of exp, and why I suggested looking at make.link("probit"). You can write your linkfun and linkinv more accurately using plogis and qlogis. On Mon, 17 Apr 2006, Mark Herzog wrote: > I was a little hesitant to post to everyone until I figured out why > there is a discrepancy in the intercept estimates when compared to the > same model run in SAS vs. R. Everything else comes out correctly, > including the other coefficient estimates... so perhaps it is just the > numerical method used. I think glm in R is using IWLS, and SAS is using ML. > > If anyone has another idea feel free to let me know. Here is my > modified binomial family, now called logexposure. Since there is no > other choice for the link for this new logexposure "family", I have just > embedded the make.link inside the logexposure family. > > Good luck and I'd appreciate any comments. > > logexposure<-function (link="logit",ExposureDays) { > logexp<-make.link("logit") > logexp$linkfun<-function(mu,expos=ExposureDays) { > log((mu^(1/expos))/(1-mu^(1/expos))) > } > logexp$linkinv<-function(eta,expos=ExposureDays) { > (exp(eta)/(1+exp(eta)))^expos > } > logexp$mu.eta<-function(eta,expos=ExposureDays) { > (expos*exp(eta*expos)*(1+exp(eta))^(-expos-1)) > } ... -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@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 and ViM
> "Sundar" == Sundar Dorai-Raj <[EMAIL PROTECTED]> > on Tue, 18 Apr 2006 10:43:30 -0500 writes: Sundar> Michael Graber wrote: >> Dear all, >> >> I'm starting to learn R, but I'm already programing for a >> while, using ViM as editor. Therefore I'd like to be able >> to use R together with ViM. >> >> My question now is, whether there are already people out >> there knowing how to do this in a similar easy way as >> with Emacs, and if those would be willing to share this >> knowledge. I did already research on the web on this >> topic, but i couldn't find satisfying answers, except one >> good looking approach on the ViM-website with a >> perl-script called funnel.pl, which I couldn't make >> running on my mac OSX 10.3.9 so far. >> >> I'd be grateful for help. >> >> Thanks, Michael Graber >> >> __ >> 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 Sundar> Try RSiteSearch('vim editor') at the R prompt. I got Sundar> 28 hits. yeah, sure. But did you look at them? I only browsed them quickly, but many don't seem to be relevant, and the few that *seemed* to be relevant, mentioned the Sciviews web page on "Editors for R" http://www.sciviews.org/_rgui/projects/Editors.html and that has one line for 'vim' which gives a link that goes to a "vimscript" page that gives the above mentioned 'funnel.pl' .. So, I think Michael is still looking for more useful hints... Martin Maechler, ETH Zurich __ 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 R/S course coming up this April in the bay area?
Please email me info on dates, etc __ 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 and ViM
This was just posted this week: https://www.stat.math.ethz.ch/pipermail/r-help/2006-April/092457.html and I think there are some relevant macros on the vim web page I use vim but do not use anything special with it. It gives syntax highlighting out of the box. On 4/18/06, Michael Graber <[EMAIL PROTECTED]> wrote: > Dear all, > > I'm starting to learn R, but I'm already programing for a while, using > ViM as editor. Therefore I'd like to be able to use R together with > ViM. > > My question now is, whether there are already people out there knowing > how to do this in a similar easy way as with Emacs, and if those would > be willing to share this knowledge. > I did already research on the web on this topic, but i couldn't find > satisfying answers, except one good looking approach on the ViM-website > with a perl-script called funnel.pl, which I couldn't make running on > my mac OSX 10.3.9 so far. > > I'd be grateful for help. > > Thanks, > Michael Graber > > __ > 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] R and ViM
yOn Tue, 18 Apr 2006, [EMAIL PROTECTED] wrote: > On 18-Apr-06 Sundar Dorai-Raj wrote: >> Michael Graber wrote: >>> I'm starting to learn R, but I'm already programing for >>> a while, using ViM as editor. Therefore I'd like to be >>> able to use R together with ViM. >>> [...] >> >> Try RSiteSearch('vim editor') at the R prompt. I got 28 hits. >> >> HTH, >> >> --sundar > > And, out of interest (being a vim addict), I did that, getting > > > RSiteSearch('vim editor') > A search query has been submitted to http://search.r-project.org > The results page should open in your browser shortly > > which it duly did. However -- and this is what's interesting -- > I entered the command in R running on my machine B, at which > I'm sitting. > > This machine B is independently displaying the Firefox browser > running on Machine C (via X Windows; Linux on both machines). > > And that's the browser where the search results appeared. > > So R running on B can hijack Firefox on C. How does it know? ?browseURL -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@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 and ViM
On 18-Apr-06 Prof Brian Ripley wrote: >> [...] >> > RSiteSearch('vim editor') >> A search query has been submitted to http://search.r-project.org >> The results page should open in your browser shortly >> >> which it duly did. However -- and this is what's interesting -- >> I entered the command in R running on my machine B, at which >> I'm sitting. >> >> This machine B is independently displaying the Firefox browser >> running on Machine C (via X Windows; Linux on both machines). >> >> And that's the browser where the search results appeared. >> >> So R running on B can hijack Firefox on C. How does it know? > > ?browseURL Wow! Wonders never cease! (Still unsure about the underlying mechanism, but then X can be a hall of mirrors to the uninitiated). Thanks for the hint! Ted. E-Mail: (Ted Harding) <[EMAIL PROTECTED]> Fax-to-email: +44 (0)870 094 0861 Date: 18-Apr-06 Time: 18:26:30 -- XFMail -- __ 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] Specifying an appropriate error term in a hierarchical regression
On 17 Apr 2006 23:55:14 +0200, Peter Dalgaard <[EMAIL PROTECTED]> wrote: > Confusing nesting and nesting... > > The / operator is designed to handle cases like this > > a b > 1 1 > 1 2 > 2 1 > 2 2 > > in which the numbering of b only makes sense within a - no connection > between b=1 when a=1 and when a=2 (think group1, member1). In this > case, the interaction a:b makes sense, but a main effect of b does > not. This makes sense. It sounds like there's no need for the "/" operator, assuming the factors are numbered appropriately. Is this correct? > It is not clear to me what B(A) is supposed to mean in Kirk's > terminology. Sounds like it could be the total sum of squares for the > b stratum, which is the effect of b ignoring a (?). SSB(A), as Kirk defines it, is "the pooled simple main effects of treatment B at each level of treatment A". Or [AB] - [A], if you prefer. I suppose my primary problem is specifying a different denominator for each level of the F test in the ANOVA. I can specify an overall error term (using "Error" in the model specification) but for the example in the book the F-test for factor A is MSA/MSB(A), while the F-test for factor B(A) is MSB(A)/MSerror. It doesn't seem like it's possible to do this using a single "aov" command; I have to run it once with each appropriate error term or do the F-test by hand. That can't be right, can it? -- Chris __ 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 debugging options
On Monday April 17 2006 21:08, Francisco J. Zagmutt wrote: > RSiteSearch("debug") or RSiteSearch("debugging") will give you a lot or > relevant information. I personally use library(debug) extensivelly and it > should do all the taks you asked about. There is a nice article describing > the debug lilbrary in the 2003/3 issue of R News > http://cran.r-project.org/doc/Rnews/Rnews_2003-3.pdf > > Cheers > > Francisco Wow! That is a great package. I think it does all I need. Is there a way to turn on debugging for all loaded functions? My source file contains many functions and I would prefer not to have to mtrace() each one. Something like >mtrace(how_do_I_get_a_list_of_all_loaded_functions) ? Larry __ 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] April Course***R/S-plus in San Francisco
Check out our upcoming 2-day "R/S-plus Fundamentals and Programming Techniques" in San Francisco: www.xlsolutions-corp.com/Rfund.htm San Franciso,April 27-28 Payment due AFTER the class Course Description: This two-day beginner to intermediate R/S-plus course focuses on a broad spectrum of topics, from reading raw data to a comparison of R and S. We will learn the essentials of data manipulation, graphical visualization and R/S-plus programming. We will explore statistical data analysis tools,including graphics with data sets. How to enhance your plots, build your own packages (librairies) and connect via ODBC,etc. We will perform some statistical modeling and fit linear regression models. Participants are encouraged to bring data for interactive sessions With the following outline: - An Overview of R and S - Data Manipulation and Graphics - Using Lattice Graphics - A Comparison of R and S-Plus - How can R Complement SAS? - Writing Functions - Avoiding Loops - Vectorization - Statistical Modeling - Project Management - Techniques for Effective use of R and S - Enhancing Plots - Using High-level Plotting Functions - Building and Distributing Packages (libraries) - Connecting; ODBC, Rweb, Orca via sockets and via Rjava Email us for group discounts. Email Sue Turner: [EMAIL PROTECTED] Phone: 206-686-1578 Visit us: www.xlsolutions-corp.com/training.htm Please let us know if you and your colleagues are interested in this classto take advantage of group discount. Register now to secure your seat! Interested in R/Splus Advanced course? email us. Cheers, Elvis Miller, PhD Manager Training. XLSolutions Corporation 206 686 1578 www.xlsolutions-corp.com [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] Compare two Power law or Exponential distributions
Dear All, I'd like to compare exponential or power-law distributions. To do so, people are often referred to the ks.test. However, I imagine ks.test wouldn't be as powerful as a test specifically designed for a distribution type. So my question is, is there a more specific test for each of these distribution? (exponential or power-law) Thanks for your hints! E __ 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] Jeffries-Matusita distance
I've calculated as follows: matusita<-function(f1,f2,lw,up){ fmatu<-function(x) ( sqrt(f1(x)) - sqrt(f2(x)) )^2 sqrt (integrate( fmatu, lower=lw, upper=up, subdivisions=100)$value) } Josué Almansa Unitat de Recerca en Serveis Sanitaris Institut Municipal d'Investigació Mèdica, IMIM c/ Doctor Aiguader, 80 08003 Barcelona e-mail: [EMAIL PROTECTED] [[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] 3D pie
You can see my opinion of 3D piecharts at http://www.burns-stat.com/pages/Tutor/spreadsheet_addiction.html Patrick Burns [EMAIL PROTECTED] +44 (0)20 8525 0696 http://www.burns-stat.com (home of S Poetry and "A Guide for the Unwilling S User") COMTE Guillaume wrote: >Hi all, > > > >Is there a way to draw 3D pie with R (like excel does)? > > > >I know how to do it in 2D, just by using pie(something)... > > > >I know it isn't the best way to represent data, but people are sometimes >more interested by the look and feel than by the accuracy of the >results... > > > >If there is no way, have you another suggestion ? (i already use >dotchart instead of pie) > > > >Thks to all of you. > >COMTE Guillaume > > > [[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] predict.nls confidence intervals
Hello- It has been several years since anyone has asked, so i am asking again- has anyone created a routine to estimate confidence intervals for predictions from nls models (ala Bates and Watts 1988)? Thanks - Alice Shelly [[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] installation of package "randomForest" failed
Hello I'd like to try out some functions in the package randomForest. Therefore, I did install this package. However, it is not possible to load the library, although I have R-Version 2.1.1 (i.e. later than 2.0.0). The commands I used and the Answers/Error from R is as follows: > install.packages("C://Programme//R//rw2011//library//randomForest_4.5-16.zip", lib="C://Programme//R//rw2011//library//randomForest",repos = NULL) package 'randomForest' successfully unpacked and MD5 sums checked updating HTML package descriptions > library(randomForest) Fehler in library(randomForest) : 'randomForest' ist kein gültiges Paket -- vor 2.0.0 installiert? I run in this problem for the first time. Until now, I used to install new packages by copying the*.zip-file to the "library"-directory in the rw2011-directory. This seam not to work anymore. So I tried using the function install.package(), but I did not success. Does anybody know the Problem? I did not find any hints in the NEWS. Thanks anyway. Ruth ** This footnote confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. www.clearswift.com __ 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] what happen?
zhang jian wrote: > *> n=PIKO[status="snag",] > Error in "[.data.frame"(PIKO, status = "snag", ) : > unused argument(s) (status ...) "=" is the assignment operator, you should use "==" as in: n=PIKO[status=="snag",] __ LLama Gratis a cualquier PC del Mundo. Llamadas a fijos y móviles desde 1 céntimo por minuto. http://es.voice.yahoo.com __ 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] installation of package "randomForest" failed
Ruth, Does your R know about the library "C://Programme//R//rw2011//library//randomForest" ? I suspect that you want lib = "C://Programme//R//rw2011//library" in your install.packages() call. Or you could just leave the 'lib=' argument at its default. Peter Ehlers Ruth Meili wrote: > Hello > > I'd like to try out some functions in the package randomForest. Therefore, > I did install this package. However, it is not possible to load the > library, although I have R-Version 2.1.1 (i.e. later than 2.0.0). The > commands I used and the Answers/Error from R is as follows: > > > > install.packages("C://Programme//R//rw2011//library//randomForest_4.5-16.zip", > > lib="C://Programme//R//rw2011//library//randomForest",repos = NULL) > package 'randomForest' successfully unpacked and MD5 sums checked > updating HTML package descriptions > > library(randomForest) > Fehler in library(randomForest) : 'randomForest' ist kein gültiges Paket -- > vor 2.0.0 installiert? > > I run in this problem for the first time. Until now, I used to install new > packages by copying the*.zip-file to the "library"-directory in the > rw2011-directory. This seam not to work anymore. So I tried using the > function install.package(), but I did not success. > > Does anybody know the Problem? I did not find any hints in the NEWS. > Thanks anyway. > Ruth > > > > > > > ** > This footnote confirms that this email message has been swept by > MIMEsweeper for the presence of computer viruses. > > www.clearswift.com > > __ > 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] store levels in a string
Il giorno mar, 18/04/2006 alle 11.29 -0500, Marc Schwartz (via MN): > > paste(levels(af), collapse = "") > [1] "ab" > > See ?paste. :) more simple: toString(levels(x)) Thanks, Cheers -- DM __ 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] nls - nonlinear regression
Hi, I am trying to run the following nonlinear regression model. > nreg <- nls(y ~ exp(-b*x), data = mydf, start = list(b = 0), alg = "default", trace = TRUE) OUTPUT: 24619327 : 0 24593178 : 0.0001166910 24555219 : 0.0005019005 24521810 : 0.001341571 24500774 : 0.002705402 24490713 : 0.004401078 24486658 : 0.00607728 24485115 : 0.007484372 24484526 : 0.008552635 24484298 : 0.009314779 24484208 : 0.009837009 24484172 : 0.01018542 24484158 : 0.01041381 24484152 : 0.01056181 24484150 : 0.01065700 24484149 : 0.01071794 24484148 : 0.01075683 24484148 : 0.01078161 24484148 : 0.01079736 24484148 : 0.01080738 24484148 : 0.01081374 Nonlinear regression modelmodel: y ~ exp(-b * x) data: mydfb 0.01081374residual sum-of-squares: 24484148 My question is how do I interpret the results of this model. > profile(nreg) 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : 24484156 : Error in prof$getProfile() : number of iterations exceeded maximum of 50 I am unable to understand the error cause. Any pointers would be of great help. Regards,Sachin - [[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] store levels in a string
On Tue, 2006-04-18 at 20:53 +0200, Daniele Medri wrote: > Il giorno mar, 18/04/2006 alle 11.29 -0500, Marc Schwartz (via MN): > > > paste(levels(af), collapse = "") > > [1] "ab" > > > > See ?paste. > > :) more simple: toString(levels(x)) > > Thanks, > > Cheers > -- > DM Actually, you don't quite get the same result: > toString(levels(af)) [1] "a, b" Note the inclusion of the ', '. See the first line in toString.default() which is: string <- paste(x, collapse = ", ") HTH, Marc __ 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] predict.nls confidence intervals
It is _just two days_ since someone asked: https://stat.ethz.ch/pipermail/r-help/2006-April/092437.html and that question was answered several times in the recent past, so I at least declined to answer it again. A quick search via RSiteSearch() shows several explanations http://finzi.psych.upenn.edu/R/Rhelp02a/archive/68606.html http://finzi.psych.upenn.edu/R/Rhelp02a/archive/35599.html http://finzi.psych.upenn.edu/R/Rhelp02a/archive/7549.html and even an (approximate) solution http://finzi.psych.upenn.edu/R/Rhelp02a/archive/42932.html If non-linearity matters, simulation can be very effective. On Tue, 18 Apr 2006, Alice Shelly wrote: > Hello- > It has been several years since anyone has asked, so i am asking again- > has anyone created a routine to estimate confidence intervals for > predictions from nls models (ala Bates and Watts 1988)? > > Thanks - > Alice Shelly > [[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 > -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@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] questioin about cluster in R
Hi,All.Sorry for the group mail. I recently met a question and I have struggled on that for a while but failed to found the solution. I have a distance matrix as below. --- 03533936 350103251 331003049 93230035 365149350 --- I want to do cluster with average method. rown<-c("A", "B", "C", "D", "E") mydistMatrix <- read.table("D:\\5.distance",row.names = rown) mydistObj<-as.dist(mydistMatrix, diag = FALSE, upper = FALSE) mycluster <- hclust(mydistObj,method="average") bmp(filename = " D:\\5_ave.bmp") plot(mycluster,hang=-1) dev.off() --- The result is something like | 20| | ___ 15| | | | | | 10| | --- | | | (intersection) | 5 | | ---| | | | | -- 0 | | | | | | EA DBC Then I want to set a threshold to cluster them. Say 5. But I don't know when A-D distance is larger than 5 or not. I can draw a line to see whether A-D distance is larger than 5 or but. When when the dataset is large, it is difficult to tell. So I wonder whether there is a way in R to display the distance value at the intersection so that we can see the exact value of it. or there is way to show or save the distance matrix after the average algorithm. Thanks a lots! Focus _ Dont just search. Find. Check out the new MSN Search! __ 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] installation of package "randomForest" failed
Ruth Meili wrote: > Hello > > I'd like to try out some functions in the package randomForest. Therefore, > I did install this package. However, it is not possible to load the > library, although I have R-Version 2.1.1 (i.e. later than 2.0.0). The > commands I used and the Answers/Error from R is as follows: > > > > install.packages("C://Programme//R//rw2011//library//randomForest_4.5-16.zip", > > lib="C://Programme//R//rw2011//library//randomForest",repos = NULL) > package 'randomForest' successfully unpacked and MD5 sums checked > updating HTML package descriptions > > library(randomForest) > Fehler in library(randomForest) : 'randomForest' ist kein gültiges Paket -- > vor 2.0.0 installiert? Where did you get the zip file from? Most easily, you can install a package by simply saying install.packages("randomForest") if you have an active internet connection. Anyway, you might want to upgrade your version of R to R-2.2.1 or even R-2.3.0 beta (to be released next monday) and help testing. Uwe Ligges > I run in this problem for the first time. Until now, I used to install new > packages by copying the*.zip-file to the "library"-directory in the > rw2011-directory. This seam not to work anymore. So I tried using the > function install.package(), but I did not success. > > Does anybody know the Problem? I did not find any hints in the NEWS. > Thanks anyway. > Ruth > > > > > > > ** > This footnote confirms that this email message has been swept by > MIMEsweeper for the presence of computer viruses. > > www.clearswift.com > > __ > 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] second try; writing user-defined GLM link function
A couple more comments: The null deviance is wrong here, as the code assumes that the link function maps constant vectors to constant vectors, which it does not here. You can circumvent that by setting an offset. Even setting dispersion = 1 I get slightly different se's. Here's a more robust way to specify the link: logexp <- function(days = 1) { linkfun <- function(mu) qlogis(mu^(1/days)) linkinv <- function(eta) plogis(eta)^days mu.eta <- function(eta) days*.Call("logit_mu_eta", eta, PACKAGE = "stats")*plogis(eta)^(days-1) valideta <- function(eta) TRUE link <- paste("logexp(", days, ")", sep="") structure(list(linkfun = linkfun, linkinv = linkinv, mu.eta = mu.eta, valideta = valideta, name = link), class = "link-glm") } and in 2.4.0 you will able to use binomial(logexp(nestdata$exposure)). On Mon, 17 Apr 2006, Mark Herzog wrote: > I was a little hesitant to post to everyone until I figured out why > there is a discrepancy in the intercept estimates when compared to the > same model run in SAS vs. R. Everything else comes out correctly, > including the other coefficient estimates... so perhaps it is just the > numerical method used. I think glm in R is using IWLS, and SAS is using ML. > > If anyone has another idea feel free to let me know. Here is my > modified binomial family, now called logexposure. Since there is no > other choice for the link for this new logexposure "family", I have just > embedded the make.link inside the logexposure family. > > Good luck and I'd appreciate any comments. > > logexposure<-function (link="logit",ExposureDays) { > logexp<-make.link("logit") > logexp$linkfun<-function(mu,expos=ExposureDays) { > log((mu^(1/expos))/(1-mu^(1/expos))) > } > logexp$linkinv<-function(eta,expos=ExposureDays) { > (exp(eta)/(1+exp(eta)))^expos > } > logexp$mu.eta<-function(eta,expos=ExposureDays) { > (expos*exp(eta*expos)*(1+exp(eta))^(-expos-1)) > } > variance <- function(mu) mu * (1 - mu) > validmu <- function(mu) all(mu > 0) && all(mu < 1) > dev.resids <- function(y, mu, wt) .Call("binomial_dev_resids", > y, mu, wt, PACKAGE = "stats") > aic <- function(y, n, mu, wt, dev) { > m <- if (any(n > 1)) > n > else wt > -2 * sum(ifelse(m > 0, (wt/m), 0) * dbinom(round(m * > y), round(m), mu, log = TRUE)) > } > initialize <- expression({ > if (NCOL(y) == 1) { > if (is.factor(y)) y <- y != levels(y)[1] > n <- rep.int(1, nobs) > if (any(y < 0 | y > 1)) stop("y values must be 0 <= y <= 1") > mustart <- (weights * y + 0.5)/(weights + 1) > m <- weights * y > if (any(abs(m - round(m)) > 0.001)) > warning("non-integer successes in a binomial glm!") > } else if (NCOL(y) == 2) { > if (any(abs(y - round(y)) > 0.001)) > warning("non-integer counts in a binomial glm!") > n <- y[, 1] + y[, 2] > y <- ifelse(n == 0, 0, y[, 1]/n) > weights <- weights * n > mustart <- (n * y + 0.5)/(n + 1) > } else stop("for the binomial family,", > " y must be a vector of 0 and 1's\n", > "or a 2 column matrix where col 1 is", > " no. successes and col 2 is no. failures") > }) > structure(list(family="logexposure", link=logexp, > linkfun=logexp$linkfun, > linkinv=logexp$linkinv, variance=variance, dev.resids=dev.resids, > aic=aic, mu.eta=logexp$mu.eta, initialize=initialize, > validmu=validmu, > valideta=logexp$valideta), class = "family") > } > > #Example > nestdata<-read.table("http://www.branta.org/ShafferChatNestData/chat.txt";) > chat.glm.logexp<-glm(survive/trials~parastat+nest_ht*patsize, > family=logexposure(ExposureDays=nestdata$expos),data=nestdata) > # if you have MASS installed > library(MASS) > chat.step<-stepAIC(chat.glm.logexp,scope=list(upper=~parastat+nest_ht*patsize,lower=~1)) > chat.step$anova > summary(chat.step) > > #Not run: > # note this compares to following results from SAS : > # Criteria For Assessing Goodness Of Fit > # > #Criterion DF ValueValue/DF > #Deviance 289193.9987 0.6713 > #Scaled Deviance 289193.9987 0.6713 > #Pearson Chi-Square 289537.8609 1.8611 > #Scaled Pearson X2289537.8609 1.8611 > #Log Likelihood -96.9994 > #Algorithm converged. > # Analysis Of Parameter Estimates > # > #Standard Wald 95% Chi- > #Parameter DF Estimate ErrorLimits Square Pr > ChiSq > ##Intercept 1 2.6973 0.2769 2.1546 3.2399 94.92 <.0001 > #parastat01 -1.0350 0
[R] R and tcltk
Hi people, I'm trying to run simple commands using tcltk in R but I got strange errors. Look at this: >library(tcltk) Loading Tcl/Tk interface ... done >tclRequire("Iwidgets") 4.0.1 >tt <- tktoplevel() >tn <- tkwidget(tt, "iwidgets::tabnotebook") Erro em structure(.External("dotTclObjv", objv, PACKAGE = "tcltk"), class = "tclObj") : [tcl] wrong # args: should be "switch ?switches? string pattern body ... ?default body?". I recently installed R 2.2.1 in suse 10. I've been using the last versions of R and I didn't have those problems with tcltk. I'll appreciate some help. Thanx Talita ___ Talita Perciano Costa Leite Mestranda em Ciência da Computação Universidade Federal de São Carlos - UFSCar Departamento de Computação Grupo de Arquitetura, Processamento de Imagens e Sinais - GAPIS __ 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 Reference Card (especially useful for Newbies)
Newbies (and others!) may find useful the R Reference Card made available by Tom Short and Rpad at http://www.rpad.org/Rpad/Rpad-refcard.pdf or through the "Contributed" link on CRAN (where some other reference cards are also linked). It categorizes and organizes a bunch of R's basic, most used functions so that they can be easily found. For example, paste() is under the "Strings" heading and expand.grid() is under "Data Creation." For newbies struggling to find the right R function as well as veterans who can't quite remember the function name, it's very handy. -- 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 __ 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] April Course***R/S-plus in San Francisco
Check out our upcoming 2-day "R/S-plus Fundamentals and Programming Techniques" in San Francisco: www.xlsolutions-corp.com/Rfund.htm San Franciso,April 27-28 Payment due AFTER the class Course Description: This two-day beginner to intermediate R/S-plus course focuses on a broad spectrum of topics, from reading raw data to a comparison of R and S. We will learn the essentials of data manipulation, graphical visualization and R/S-plus programming. We will explore statistical data analysis tools,including graphics with data sets. How to enhance your plots, build your own packages (librairies) and connect via ODBC,etc. We will perform some statistical modeling and fit linear regression models. Participants are encouraged to bring data for interactive sessions With the following outline: - An Overview of R and S - Data Manipulation and Graphics - Using Lattice Graphics - A Comparison of R and S-Plus - How can R Complement SAS? - Writing Functions - Avoiding Loops - Vectorization - Statistical Modeling - Project Management - Techniques for Effective use of R and S - Enhancing Plots - Using High-level Plotting Functions - Building and Distributing Packages (libraries) - Connecting; ODBC, Rweb, Orca via sockets and via Rjava Email us for group discounts. Email Sue Turner: [EMAIL PROTECTED] Phone: 206-686-1578 Visit us: www.xlsolutions-corp.com/training.htm Please let us know if you and your colleagues are interested in this classto take advantage of group discount. Register now to secure your seat! Interested in R/Splus Advanced course? email us. Cheers, Elvis Miller, PhD Manager Training. XLSolutions Corporation 206 686 1578 www.xlsolutions-corp.com [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] R debugging options
Dear Larry, I'm not aware of an existing function that lists functions, but here's a simple solution: listFunctions <- function(all.names=FALSE, envir=.GlobalEnv){ # all.names=TRUE: include names beginning with "." # envir: environment to search Objects <- objects(envir, all.names=all.names) if (length(Objects) == 0) Objects else names(which(sapply(Objects, function(object) is.function(eval(parse(text=object), envir=envir) } Getting mtrace() to use the function names returned by listFunctions() is a bit tricky, because of the way mtrace() evaluates its arguments. You could do something like the following: for(f in listFunctions()) mtrace(char.fname=f) Perhaps someone else knows of an existing or better solution. 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 Larry Howe > Sent: Tuesday, April 18, 2006 12:46 PM > To: r-help@stat.math.ethz.ch > Subject: Re: [R] R debugging options > > On Monday April 17 2006 21:08, Francisco J. Zagmutt wrote: > > RSiteSearch("debug") or RSiteSearch("debugging") will give > you a lot > > or relevant information. I personally use library(debug) > extensivelly > > and it should do all the taks you asked about. There is a > nice article > > describing the debug lilbrary in the 2003/3 issue of R News > > http://cran.r-project.org/doc/Rnews/Rnews_2003-3.pdf > > > > Cheers > > > > Francisco > > Wow! That is a great package. I think it does all I need. > > Is there a way to turn on debugging for all loaded functions? > My source file contains many functions and I would prefer not > to have to mtrace() each one. > Something like > > >mtrace(how_do_I_get_a_list_of_all_loaded_functions) > > ? > > Larry > > __ > 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] R and ViM
(Ted Harding) <[EMAIL PROTECTED]> writes: > On 18-Apr-06 Prof Brian Ripley wrote: > >> [...] > >> > RSiteSearch('vim editor') > >> A search query has been submitted to http://search.r-project.org > >> The results page should open in your browser shortly > >> > >> which it duly did. However -- and this is what's interesting -- > >> I entered the command in R running on my machine B, at which > >> I'm sitting. > >> > >> This machine B is independently displaying the Firefox browser > >> running on Machine C (via X Windows; Linux on both machines). > >> > >> And that's the browser where the search results appeared. > >> > >> So R running on B can hijack Firefox on C. How does it know? > > > > ?browseURL > > Wow! Wonders never cease! (Still unsure about the underlying > mechanism, but then X can be a hall of mirrors to the uninitiated). > > Thanks for the hint! > Ted. Actually, I think this belongs in a group of _mis_features in the current user interface landscape, along with saved sessions & the like. It sounds smart, right until you want to view a file on machine B, say file://foo, which machine C and hence Firefox has no knowledge about. You then realize that the only way you are going to view that file is to shut down Firefox on C and start a new one on B. Attempting to start one on B, as in "firefox file://foo" will just send a message to the one on C telling it to display file://foo... I usually get bitten by this in the reverse situation: remote login to work machine, run R, do help.start(), and "Firefox can't find the file at /tmp/RtmpZVbVef/.R/doc/html/index.html". It sort of indicates missing perspective on the part of the developers: They have forgotten about the fact that browsers might be used for information *not* on the Internet. (SAS tries to get around the problem of non-local access to its help pages by setting up a local HTTP server, only to be foiled by firewalling...) -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ 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] Different results under Windows and Linux
Hi everyone, disturbingly, a PCA yields different results when I run the analysis under Windows and Linux. I open the same workspace (which is on an independent partition) and perform the exact same function (by using the history file). Interestingly, PC1 is the same, but PC2 comes out totally different. Thanks! Kamila Windows: platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.1 year 2005 month12 day 20 svn rev 36812 language R Linux (Suse 10, Kernel 2.6.13-15.8-default) platform i686-pc-linux-gnu arch i686 os linux-gnu system i686, linux-gnu status major2 minor2.1 year 2005 month12 day 20 svn rev 36812 language R __ 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 and ViM
Hmm, how timely. I posted yesterday my solution to integrate R and vim. The message is in my sent box but I don't think it showed up in the list... Here it is again: Hi All, If you use vim to edit R code, you may be interested in this. I have put together a personalized syntax file, some code templates, and a way to send code from Vim to R using autoHotKeys (windows). http://www.andrew.cmu.edu/user/jquesada/RvimSuite/instructions.html Actually, the little autoHotKeys can be useful even if you don't use vim just to send the example R code from the help pages to the console. Best wishes, -Jose PS: @list moderators Any idea why my message (from a @gmail account) appeared in the sent box but never on the list? On 4/18/06, Michael Graber <[EMAIL PROTECTED]> wrote: > Dear all, > > I'm starting to learn R, but I'm already programing for a while, using > ViM as editor. Therefore I'd like to be able to use R together with > ViM. > > My question now is, whether there are already people out there knowing > how to do this in a similar easy way as with Emacs, and if those would > be willing to share this knowledge. > I did already research on the web on this topic, but i couldn't find > satisfying answers, except one good looking approach on the ViM-website > with a perl-script called funnel.pl, which I couldn't make running on > my mac OSX 10.3.9 so far. > > I'd be grateful for help. > > Thanks, > Michael Graber > > __ > 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 > -- Cheers, -Jose -- Jose Quesada, PhD. [EMAIL PROTECTED] Dept. of Psychology http://www.andrew.cmu.edu/~jquesada Sussex University Brighton, UK __ 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 and ViM
"Jose Quesada" <[EMAIL PROTECTED]> writes: > Hmm, how timely. > > I posted yesterday my solution to integrate R and vim. The message is > in my sent box but I don't think it showed up in the list... Here it > is again: > > Hi All, > > If you use vim to edit R code, you may be interested in this. > I have put together a personalized syntax file, some code templates, > and a way to send code from Vim to R using autoHotKeys (windows). > > http://www.andrew.cmu.edu/user/jquesada/RvimSuite/instructions.html > > Actually, the little autoHotKeys can be useful even if you don't use > vim just to send the example R code from the help pages to the > console. > > Best wishes, > -Jose > > PS: > @list moderators > Any idea why my message (from a @gmail account) appeared in the sent > box but never on the list? I saw it the first time... Could it be that your mail reader is set up to ignore mail from yourself? Rather than resending stuff, it is preferable to check the archives https://stat.ethz.ch/pipermail/r-help/2006-April/date.html https://stat.ethz.ch/pipermail/r-help/2006-April/092457.html which are easily reachable via > > R-help@stat.math.ethz.ch mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ 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] Different results under Windows and Linux
Thanks Bert, I did not describe the problem accurately enough. Actually, by now, I found out that floating-point imprecision is the reason for my confusion. The problem does not arise with the prcomp() function but rather with my data preprocessing: > means=apply(dev,2,mean,na.rm=T) > means windows 371.19100102494, 432.458908438674 etc. linux 371.191001024941,432.458908438674 etc. > stdev=apply(dev,2,sd,na.rm=T) > stdev windows 872.234567580374, 1324.05572667875 etc. linux 872.234567580372, 1324.05572667875 etc. Thanks! Kamila Berton Gunter wrote: >Please provide better documentation to help diagnose the problem. What are >the PC's ? > >-- 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 Kamila Naxerova >>Sent: Tuesday, April 18, 2006 3:12 PM >>To: r-help@stat.math.ethz.ch >>Subject: [R] Different results under Windows and Linux >> >>Hi everyone, >> >>disturbingly, a PCA yields different results when I run the analysis >>under Windows and Linux. >>I open the same workspace (which is on an independent partition) and >>perform the exact same function (by using the history file). >>Interestingly, PC1 is the same, but PC2 comes out totally different. >> >>Thanks! >>Kamila >> >>Windows: >>platform i386-pc-mingw32 >>arch i386 >>os mingw32 >>system i386, mingw32 >>status >>major2 >>minor2.1 >>year 2005 >>month12 >>day 20 >>svn rev 36812 >>language R >> >> >>Linux (Suse 10, Kernel 2.6.13-15.8-default) >>platform i686-pc-linux-gnu >>arch i686 >>os linux-gnu >>system i686, linux-gnu >>status >>major2 >>minor2.1 >>year 2005 >>month12 >>day 20 >>svn rev 36812 >>language R >> >>__ >>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] How to efficiently extract or construct the residual covariance matrix from lme()?
Dear R-community, I'm trying to get the estimated residual covariance matrices from an lme object. If we write the model as: Y = X \beta + Z b + \epsilon and assume that b ~ N(0, P) and \epsilon ~ N(0, \Sigma), where P is non-diagonal and \Sigma might have correlation and weights components, then I'm looking for efficient estimates of \Sigma and ZPZ' + \Sigma I can find P easily enough, but I'm wondering if there's an easy way to get at Z and \Sigma. Also I can move to lmer() if that simplifies the problem. Cheers Andrew -- Andrew Robinson Department of Mathematics and StatisticsTel: +61-3-8344-9763 University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599 Email: [EMAIL PROTECTED] http://www.ms.unimelb.edu.au __ 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] about "intersection set" and "union set"
Hello sir: group1:1,2,3,4 group2:1,3,4,5 group3:2,4,8,9 .. ... group1000:9,3,8,2 I wanna get the "intersection set" and "union set". I've tried command "setdiff"and "union",but only two groups is permited.How can I deal with multi groups to find the "intersection set" and "union set"? Thanks a lot! My best -- *** Xin Meng Capitalbio Corporation National Engineering Research Center for Beijing Biochip Technology BioPharma-informatics & Software Dept. Research Engineer Tel: +86-10-80715888/80726868-6438 Fax: +86-10-80726790 [EMAIL PROTECTED] Address:18 Life Science Parkway, Changping District, Beijing 102206, China __ 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] about "intersection set" and "union set"
This was just discussed recently. See the following thread: http://tolstoy.newcastle.edu.au/R/help/06/04/24923.html On 4/18/06, XinMeng <[EMAIL PROTECTED]> wrote: > Hello sir: > > group1:1,2,3,4 > group2:1,3,4,5 > group3:2,4,8,9 > .. ... > group1000:9,3,8,2 > > I wanna get the "intersection set" and "union set". I've tried command > "setdiff"and "union",but only two groups is permited.How can I deal with > multi groups to find the "intersection set" and "union set"? > > Thanks a lot! > My best > > > > > > -- > *** > Xin Meng > Capitalbio Corporation > National Engineering Research Center > for Beijing Biochip Technology > BioPharma-informatics & Software Dept. > Research Engineer > Tel: +86-10-80715888/80726868-6438 > Fax: +86-10-80726790 > [EMAIL PROTECTED] > Address:18 Life Science Parkway, > Changping District, Beijing 102206, China > > > > __ > 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] isoMDS and 0 distances
Hi, I'm trying to do a non-metric multidimensional scaling using isoMDS. However, I have some '0' distances in my data, and I'm not sure how to deal with them. I'd rather not drop rows from the original data, as I am comparing several datasets (morphology and molecular data) for the same individuals, and it's interesting to see how much morphological variation can be associated with an identical genotype. I've tried replacing the 0's with NA, but the isoMDS appears to stop on the first iteration and the stress does not improve: distA # A dist object with 13695 elements, 4 of which == 0 cmdsA <- cmdscale(distA, k=2) distB <- distA distB[which(distB==0)] <- NA isoA <- isoMDS(distB, cmdsA) initial value 21.835691 final value 21.835691 converged The other approach I've tried is replacing the 0's with small numbers. In this case isoMDS does reduce the stress values. min(distA[which(distA>0)]) [1] 0.02325581 distC <- distA distC[which(distC==0)] <- 0.001 isoC <- isoMDS(distC) initial value 21.682854 iter 5 value 16.862093 iter 10 value 16.451800 final value 16.339224 converged So my questions are: what am I doing wrong in the first example? Why does isoMDS converge without doing anything? Is replacing the 0's with small numbers an appropriate alternative? Thanks for your time, Tyler R 2.2.1 __ 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] second try; writing user-defined GLM link function
It's probable that Roger's pdf that he provided is much more appropriate (or perhaps may have a good reference). However, if interested, the specific example used for my code, is from the following paper (and is now widely being used in many avian nesting studies, but I leave to others to decide if useful in a much broader context): Shaffer, T. 2004. A unifying approach to analyzing nest success. Auk 121(2): 526-540. In this paper Shaffer develops the link, but only cites McCullugh/Nelder, and Hosmer/Lemeshow (which I don't think discuss any other links for the binomial family besides those already available in R) On his website ( http://www.npwrc.usgs.gov/resource/birds/nestsurv/nestsurv.htm ), Shaffer has some code examples in SAS (with the data embedded in his example code). My previous email, with code and example, provides a direct link to a datafile with that same data. I am fairly certain he would be willing to offer that data for use, if desired. Mark Mark Herzog, Ph.D. Program Leader, San Francisco Bay Research Wetland Division, PRBO Conservation Science 4990 Shoreline Highway 1 Stinson Beach, CA 94970 (415) 893-7677 x308 [EMAIL PROTECTED] Prof Brian Ripley wrote: > Is this something that is fairly widely useful? If so, it is something > that we could add to the stats package. (We ought in any case to make it > easier to extend glm families, and this could be an example. If so, I'd > like a reference for its use.) > > There is a good reason why logit is implmented as > >> make.link("logit") > $linkfun > function (mu) > .Call("logit_link", mu, PACKAGE = "stats") > > > $linkinv > function (eta) > .Call("logit_linkinv", eta, PACKAGE = "stats") > > > $mu.eta > function (eta) > .Call("logit_mu_eta", eta, PACKAGE = "stats") > > > $valideta > function (eta) > TRUE > > > which is that these functions protect against underflow/overflow of exp, > and why I suggested looking at make.link("probit"). > > You can write your linkfun and linkinv more accurately using plogis and > qlogis. > > > On Mon, 17 Apr 2006, Mark Herzog wrote: > >> I was a little hesitant to post to everyone until I figured out why >> there is a discrepancy in the intercept estimates when compared to the >> same model run in SAS vs. R. Everything else comes out correctly, >> including the other coefficient estimates... so perhaps it is just the >> numerical method used. I think glm in R is using IWLS, and SAS is using ML. >> >> If anyone has another idea feel free to let me know. Here is my >> modified binomial family, now called logexposure. Since there is no >> other choice for the link for this new logexposure "family", I have just >> embedded the make.link inside the logexposure family. >> >> Good luck and I'd appreciate any comments. >> >> logexposure<-function (link="logit",ExposureDays) { >> logexp<-make.link("logit") >> logexp$linkfun<-function(mu,expos=ExposureDays) { >> log((mu^(1/expos))/(1-mu^(1/expos))) >> } >> logexp$linkinv<-function(eta,expos=ExposureDays) { >> (exp(eta)/(1+exp(eta)))^expos >> } >> logexp$mu.eta<-function(eta,expos=ExposureDays) { >> (expos*exp(eta*expos)*(1+exp(eta))^(-expos-1)) >> } > > ... > __ 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 and ViM
[Michael Graber] >[...] I'd like to be able to use R together with ViM. [...] My >question now is, whether there are already people out there knowing how >to do this in a similar easy way as with Emacs [...] I've been an Emacs user for a very long time, and then, switched to Vim. See http://pinard.progiciels-bpi.ca/opinions/editors.html, if you feel curious, for a few personal thoughts on Emacs. For R, I tried sticking to a mere interactive shell, taking advantage of the "GNU readline" interface built into R, with Vim as an external editor. For sending R code from Vim to R, one merely selects the code to send within Vim using the mouse, and paste it directly with the mouse in the interactive shell window running R. Simple and comfortable! :-) Emacs offers ESS, which has many interesting features. However, despite quite attractive, it did not fully seduce me: a bit because I try to avoid returning to Emacs keystroke habits, a bit because ESS is heavy weighted compared to Vim + R-in-a-shell solution, a bit because ESS adds distracting idiosyncrasies, like scrolling differently or opening extra windows at times. R already offers enough options I could customize if I want to read help in a browser or a pager, and at good speed. (Of course, if you use an heavy browser, you feel it; but "links -g" is OK!) An ESS nicety that my current setup does not really replace is the automatic highlighting or R output. One of the advantages of this output highlighting is visually spotting R requests and replies. As a compromise, I'm using this bit of a kludge in my Rprofile file: if (interactive()) { local({ options(editor='vim -c "set ft=r"') if (Sys.getenv('TERM') %in% c('rxvt', 'xterm')) { onglet = 2 options(prompt=paste(sep='', formatC('', width=80-onglet), '\033[;30;45m', formatC('', width=onglet), '\033[0m\n', options('prompt'))) } }) } The "set ft=r" bit ensures proper highlighting and coloration within Vim, whenever edit() or fix() are used. Here "vim" could be replaced by "gvim" or "gvim -f", say. (In my Vim configuration, "vim" uses the GUI automatically if started within X; or uses the console mode otherwise.) Then, the R prompt is modified to visually mark each request-reply interaction with a white separating line holding a small violet marker at the right. It works nicely for me in almost all circumstances (there are a few, uncommon exceptions). Usual scrolling of the shell window allows me to quickly find R commands and replies, even if much less colourful than with ESS. I'm ready to pay that price for simplicity. A last trick which is convenient in my case. My X window manager allows customization of keystrokes. (I'm using Openbox, but surely many other window manages offer that possibility too.) For all 26 of Ctrl-Alt-Letter, the same small "openbox-helper" (Python) script of mine is called with the Letter given as an option, which may launch applications in turn. This is how Ctrl-Alt-R opens a shell window running R, and Ctrl-Alt-M opens a shell window running Maxima. In both these shells, Ctrl-D closes the application and the window. This is convenient for quick mathematical jobs, and quite in the spirit of Vim (fast and easy start/exit, instead of long running like Emacs). -- François Pinard http://pinard.progiciels-bpi.ca __ 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] Thousands Separator
I am a new R user, and am struggling with how to add a comma to the tick mark labels in a bwplot or a boxplot. I think the default values of the labels are fine, I just want them to appear with commas. If I use the function boxplot, I can then use axis(2,labels=format(axTicks(2),big.mark=",")) but that just overlays them on the existing labels which I don't know how to turn off. This method does not appear to work at all with bwplot, but I am not sure why. Any help would be appreciated. Thanks, David --- David Feldman The Taylor Feldman Group, LLC www.taylorfeldman.com (530) 400-5470 __ 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-pkgs] Version 1.0 of hapassoc now available from CRAN
Version 1.0 of hapassoc now available from CRAN hapassoc is an R package for likelihood inference of trait associations with SNP haplotypes and other attributes using the EM Algorithm. Recent changes include the addition of anova and logLik methods for the class hapassoc, to allow users to perform likelihood ratio tests of haplotype effects. Other changes include bug-fixes and improvements to the documentation, such as the addition of a package vignette. See the exerpt from the ChangeLog for more details. ChangeLog 2006-04-04 Version 1.0 * hapassoc(): the "baseline" argument is documented to have default equal to the most common haplotype, but the code to implement this default was lost and needed to be replaced. * hapassoc(): added a "verbose" flag. Default is verbose=FALSE. If TRUE users see the iteration number and value of the convergence criterion at each iteration of the EM algorithm. * pre.hapassoc(): added a "verbose" flag. Default is verbose=TRUE. If TRUE users see a list of the SNP genotypes used to form haplotypes and a list of the other "non-haplotype" variables. * Package vignette "hapassoc" added. After loading the package, type vignette("hapassoc") to view. 2006-03-22 * Overall addition of the log-likelihood functions * hapassoc(): function now returns log-likelihood and model * logLik.hapassoc(): New function to extract the log-likelihood from a hapassoc object * anova.hapassoc(): New function to perform likelihood ratio test on two hapassoc objects. 2006-02-02 Minor changes: * EMvar(): fixed a bug occurring when all haplotype phases are known. * RecodeHaplos(): fixed a bug where a single column of non-haplotype data in a non-allelic data set was losing its name. * hapassoc(): Change "..." argument of hapassoc to "start". Previously the only intended use of "..." was to allow the user to pass in "start" for starting values to the glm function, rather than to allow the user to pass in other optional arguments to glm. We have now made this more explicit by making this argument more specific. ___ R-packages mailing list [EMAIL PROTECTED] https://stat.ethz.ch/mailman/listinfo/r-packages __ 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 and ViM
> "PD" == Peter Dalgaard <[EMAIL PROTECTED]> > on 19 Apr 2006 01:06:02 +0200 writes: PD> "Jose Quesada" <[EMAIL PROTECTED]> writes: >> Hmm, how timely. >> >> I posted yesterday my solution to integrate R and >> vim. The message is in my sent box but I don't think it >> showed up in the list... Here it is again: >> >> Hi All, >> >> If you use vim to edit R code, you may be interested in >> this. I have put together a personalized syntax file, >> some code templates, and a way to send code from Vim to R >> using autoHotKeys (windows). >> >> http://www.andrew.cmu.edu/user/jquesada/RvimSuite/instructions.html >> >> Actually, the little autoHotKeys can be useful even if >> you don't use vim just to send the example R code from >> the help pages to the console. >> >> Best wishes, -Jose >> >> PS: @list moderators Any idea why my message (from a >> @gmail account) appeared in the sent box but never on the >> list? PD> I saw it the first time... Could it be that your mail PD> reader is set up to ignore mail from yourself? PD> Rather than resending stuff, it is preferable to check PD> the archives PD> https://stat.ethz.ch/pipermail/r-help/2006-April/date.html PD> https://stat.ethz.ch/pipermail/r-help/2006-April/092457.html PD> which are easily reachable via >> > R-help@stat.math.ethz.ch mailing list > >> https://stat.ethz.ch/mailman/listinfo/r-help Indeed. Please do check the archives. Now back to the subject: Jose, I think your main contribution is based on "autoHotKeys" and that only works on Windoze, right? Michael explicitly mentioned he's working in Mac OS X. Martin __ 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] second try; writing user-defined GLM link function
Excellent! So to provide the final summary for everyone's sake, based on Dr. Ripley's revisions, and now a revised logexposure "family" function to use with the improved link function (until 2.4.0 makes it simpler) : # Logistical Exposure Link Function # See Shaffer, T. 2004. A unifying approach to analyzing nest success. # Auk 121(2): 526-540. logexp <- function(days = 1) { linkfun <- function(mu) qlogis(mu^(1/days)) linkinv <- function(eta) plogis(eta)^days mu.eta <- function(eta) days*.Call("logit_mu_eta", eta, PACKAGE = "stats")*plogis(eta)^(days-1) valideta <- function(eta) TRUE link <- paste("logexp(", days, ")", sep="") structure(list(linkfun = linkfun, linkinv = linkinv, mu.eta = mu.eta, valideta = valideta, name = link), class = "link-glm") } # Modified binomial family function (that allows logexp link function) logexposure<-function (link="logexp",ExposureDays) { variance <- function(mu) mu * (1 - mu) validmu <- function(mu) all(mu > 0) && all(mu < 1) dev.resids <- function(y, mu, wt) .Call("binomial_dev_resids", y, mu, wt, PACKAGE = "stats") aic <- function(y, n, mu, wt, dev) { m <- if (any(n > 1)) n else wt -2 * sum(ifelse(m > 0, (wt/m), 0) * dbinom(round(m * y), round(m), mu, log = TRUE)) } initialize <- expression({ if (NCOL(y) == 1) { if (is.factor(y)) y <- y != levels(y)[1] n <- rep.int(1, nobs) if (any(y < 0 | y > 1)) stop("y values must be 0 <= y <= 1") mustart <- (weights * y + 0.5)/(weights + 1) m <- weights * y if (any(abs(m - round(m)) > 0.001)) warning("non-integer successes in a binomial glm!") } else if (NCOL(y) == 2) { if (any(abs(y - round(y)) > 0.001)) warning("non-integer counts in a binomial glm!") n <- y[, 1] + y[, 2] y <- ifelse(n == 0, 0, y[, 1]/n) weights <- weights * n mustart <- (n * y + 0.5)/(n + 1) } else stop("for the binomial family,", " y must be a vector of 0 and 1's\n", "or a 2 column matrix where col 1 is", " no. successes and col 2 is no. failures") }) structure(list(family="binomial", link=logexp(ExposureDays), linkfun=logexp(ExposureDays)$linkfun, linkinv=logexp(ExposureDays)$linkinv, variance=variance, dev.resids=dev.resids, aic=aic, mu.eta=logexp(ExposureDays)$mu.eta, initialize=initialize, validmu=validmu, valideta=logexp$valideta), class = "family") } #Example nestdata<-read.table("http://www.branta.org/ShafferChatNestData/chat.txt";) chat.glm.logexp<-glm(survive/trials~parastat+nest_ht*patsize,family=logexposure(ExposureDays=nestdata$expos),data=nestdata) # if you have MASS installed library(MASS) chat.step<-stepAIC(chat.glm.logexp,scope=list(upper=~parastat+nest_ht*patsize,lower=~1)) chat.step$anova summary(chat.step) Mark Herzog, Ph.D. Program Leader, San Francisco Bay Research Wetland Division, PRBO Conservation Science 4990 Shoreline Highway 1 Stinson Beach, CA 94970 (415) 893-7677 x308 [EMAIL PROTECTED] Prof Brian Ripley wrote: > A couple more comments: > > The null deviance is wrong here, as the code assumes that the link > function maps constant vectors to constant vectors, which it does not > here. You can circumvent that by setting an offset. > > Even setting dispersion = 1 I get slightly different se's. > > Here's a more robust way to specify the link: > > logexp <- function(days = 1) > { > linkfun <- function(mu) qlogis(mu^(1/days)) > linkinv <- function(eta) plogis(eta)^days > mu.eta <- function(eta) > days*.Call("logit_mu_eta", eta, PACKAGE = > "stats")*plogis(eta)^(days-1) > valideta <- function(eta) TRUE > link <- paste("logexp(", days, ")", sep="") > structure(list(linkfun = linkfun, linkinv = linkinv, >mu.eta = mu.eta, valideta = valideta, name = link), > class = "link-glm") > } > > and in 2.4.0 you will able to use binomial(logexp(nestdata$exposure)). > __ 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] Thousands Separator
David Feldman wrote: > I am a new R user, and am struggling with how to add a comma to the > tick mark labels in a bwplot or a boxplot. > > I think the default values of the labels are fine, I just want them > to appear with commas. > > If I use the function boxplot, I can then use >axis(2,labels=format(axTicks(2),big.mark=",")) > but that just overlays them on the existing labels which I don't know > how to turn off. Specify in your plot function: yaxt="n" (see ?par) > This method does not appear to work at all with bwplot, but I am not > sure why. Because it is not a base graphics function but a lattice graphics. Uwe Ligges > Any help would be appreciated. > > Thanks, > > David > > --- > David Feldman > The Taylor Feldman Group, LLC > www.taylorfeldman.com > (530) 400-5470 > > __ > 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] isoMDS and 0 distances
Short answer: you cannot compare distances including NAs, so there is no way to find a monotone mapping of distances. If the data really are identical for two rows, you can easily drop one of them whilst doing MDS, and then assign the position found for one to the other. On Tue, 18 Apr 2006, Tyler Smith wrote: > Hi, > > I'm trying to do a non-metric multidimensional scaling using isoMDS. > However, I have some '0' distances in my data, and I'm not sure how to > deal with them. I'd rather not drop rows from the original data, as I am > comparing several datasets (morphology and molecular data) for the same > individuals, and it's interesting to see how much morphological > variation can be associated with an identical genotype. > > I've tried replacing the 0's with NA, but the isoMDS appears to stop on > the first iteration and the stress does not improve: > > distA # A dist object with 13695 elements, 4 of which == 0 > cmdsA <- cmdscale(distA, k=2) > > distB <- distA > distB[which(distB==0)] <- NA > > isoA <- isoMDS(distB, cmdsA) > initial value 21.835691 > final value 21.835691 > converged > > The other approach I've tried is replacing the 0's with small numbers. > In this case isoMDS does reduce the stress values. > > min(distA[which(distA>0)]) > [1] 0.02325581 > > distC <- distA > distC[which(distC==0)] <- 0.001 > isoC <- isoMDS(distC) > initial value 21.682854 > iter 5 value 16.862093 > iter 10 value 16.451800 > final value 16.339224 > converged > > So my questions are: what am I doing wrong in the first example? Why > does isoMDS converge without doing anything? Is replacing the 0's with > small numbers an appropriate alternative? > > Thanks for your time, > > Tyler > R 2.2.1 > > __ > 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 > -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@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