Re: [R] http proxies: setting and unsetting
One other point. If you find you need to set a system or user environment variable then microsoft has a free tool called setx.exe that you can find here: http://support.microsoft.com/kb/927229 You can do this from within R using system(). On 5/30/07, Gabor Grothendieck [EMAIL PROTECTED] wrote: On 5/29/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: OK, I think I get that... do you know which namespace the Sys.setenv() function affects? Do you know if there are functions in R that can alter the user/system/process environment variables? Use the R Sys.getenv() command to get the process environment variables. To get user and system environment variables, from the Desktop right click on My Computer and choose Properties. Then choose the Advanced tab and click on the Environment Variables button near the bottom of the window that appears. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] [R-pkgs] new packages psyphy and MLDS
New packages psyphy and MLDS are available on CRAN: psyphy ncludes an assortment of functions useful in analyzing data from pyschophysical experiments. It includes functions for calculating d' from several different experimental designs, links for mafc to be used with the binomial family in glm (and possibly other contexts) and selfStart functions for estimating gamma values for CRT (and possibley other RGB) screen calibration data. MLDS implements analyses for Maximum Likelihood Difference Scaling. Difference scaling is a method for scaling perceived super-threshold differences. The package contains functions that allow the user to fit the resulting data by maximum likelihood and to test the internal validity of the estimated scale. There are also example functions that might be used to design and run a difference scaling experiment, Any suggestioins, criticisms, bug-reports, etc. are always welcome. Best, Ken Knoblauch -- Ken Knoblauch Inserm U846 Institut Cellule Souche et Cerveau Département Neurosciences Intégratives 18 avenue du Doyen Lépine 69500 Bron France tel: +33 (0)4 72 91 34 77 fax: +33 (0)4 72 91 34 61 portable: +33 (0)6 84 10 64 10 http://www.pizzerialesgemeaux.com/u846/ ___ 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 and provide commented, minimal, self-contained, reproducible code.
[R] [R-pkgs] DPpackage - New version
Dear List: I have uploaded version 1.0-4 of DPpackage on CRAN. Since the first version (1.0-0), I have not communicated the improvements of the package. I'll use this email to summarize its current status. The name of the package is motivated by the Dirichlet process. However, DPpackage tries to be a general package for Bayesian nonparametric and semi-parametric data analysis. So far, the package includes models based on Dirichlet processes, Dirichlet process mixtures of normals, Polya trees, and Random Bernstein polynomials. A list of current functions is given next: 1) Density estimation: DPdensity (using DPM of normals), PTdensity (using Mixtures of Polya Trees), and BDPdensity (using Bernstein-Dirichlet prior). The first two functions allow uni- and multi-variate analysis. 2) Nonparametric random effects distributions in mixed effects models: 2.1) DPlmm and DPMlmm, using a DP/MDP and DPM of normals prior, respectively, for the linear mixed effects model. 2.2) DPglmm and DPMglmm, using a DP/MDP and DPM of normals prior, respectively, for generalized linear mixed effects models, respectively. The sampling(link) considered by these functions are binomial(logit,probit), poisson(log) and gamma(log). 2.3) DPolmm and DPMolmm, using a DP/MDP and DPM of normals prior, respectively, for the probit-ordinal mixed effects models. 2.4) DPrasch and FPTrasch, using a DP/MDP and finite PT/MPT (mixture of Polya Trees) prior for the Rasch model with binary sampling distribution, respectively. 2.5) DPraschpoisson and FPTraschpoisson. The same as before (2.4) but with a Poisson sampling. 2.6) DPmeta and DPMmeta for the random (mixed) effects meta-analysis models, using a DP/MDP and DPM of normals prior, respectively. 3) Binary regression with nonparametric link: 3.1) CSDPbinary, using Newton, Czado and Chappell (1996)'s centrally standardized DP prior. 3.2) DPbinary, using the regular DP prior for the inverse of the link function. 3.3) FPTbinary, using a finite PT prior for the inverse of the link function. 4) AFT model for interval-censored data: 4.1) DPsurvint, using a MDP prior for the baseline distribution. 5) ROC curve estimation: 5.1) DProc, using DPM of normals. 6) Linear model with a nonparametric for the error distribution: 6.1) PTlm, using MPT. 7) DP prior elicitation: 7.1) DPelicit, using the exact and approximated formulas for the mean and variance of the number of clusters given the total mass parameter and the number of subjects. Tim Hanson and Fernando Quintana have made contributions to the current version. I would also like to thank George Karabatsos for his input to the current status of the package and Peter Mueller for actively promoting the package. Various other improvements have been motivated by questions asked by many people around the world. I would like to thank all of them too. I welcome anyone who sends comments, suggestions, remarks, and particularly those who find bugs or mistakes in any part of the package or its documentation. DPpackage is an open source program for Bayesian nonparametric developments. All contributions are welcome. Best regards, Alejandro. Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm ___ 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 and provide commented, minimal, self-contained, reproducible code.
[R] functions without arguments
hello, I wanna know if it's possible to write functions without argument in order to understand that I write down an example f - function(){ for (i in 1:length(C[[1]]) { print(10*C[[1]][i]) } } I know that there are errors in this syntax _ [[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 and provide commented, minimal, self-contained, reproducible code.
[R] Re : functions without arguments
there are R function with zero args : plot.new,frame, colors. In your case make sure that you list C is present in environments that you use. Justin BEM Elève Ingénieur Statisticien Economiste BP 294 Yaoundé. Tél (00237)9597295. - Message d'origine De : elyakhlifi mustapha [EMAIL PROTECTED] À : R-help@stat.math.ethz.ch Envoyé le : Mercredi, 30 Mai 2007, 8h37mn 36s Objet : [R] functions without arguments hello, I wanna know if it's possible to write functions without argument in order to understand that I write down an example f - function(){ for (i in 1:length(C[[1]]) { print(10*C[[1]][i]) } } I know that there are errors in this syntax _ [[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 and provide commented, minimal, self-contained, reproducible code. ls.. et vos réactions ! [[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 and provide commented, minimal, self-contained, reproducible code.
[R] Factor function: odd behavior when labels argument contains duplicates?
Hi all. I think it would be nice to be able to combine levels of a factor on creation a la x - rep(0:5,5) y - factor(x,levels=0:5,labels=c('1','1',2:5)) ## (1) y [1] 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 Levels: 1 1 2 3 4 5 I thougt this would (should?) create a factor with 5 levels, ie combining 0 and 1 in x into one level in y. I find it hard to predict the behavior of the following lines: table(factor(y)) ## Odd ? 1 1 2 3 4 5 10 0 5 5 5 5 table(factor(factor(y))) ## This is what I want 1 2 3 4 5 10 5 5 5 5 It also seems strange that this should be the way to go: levels(y) - levels(y) ## Does what I want following (1) even tough it appear to be a void statement? y y [1] 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 Levels: 1 2 3 4 5 Am I missing an important point about factors or the factor function? steen Steen Ladelund, statistician +4543233275 [EMAIL PROTECTED] Research Center for Prevention and Health Glostrup University Hospital, Denmark www.fcfs.kbhamt.dk __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Re : functions without arguments
You can see Lexical Scope and Statistical Computing, Robert Gentleman; Ross Ihaka Justin BEM Elève Ingénieur Statisticien Economiste BP 294 Yaoundé. Tél (00237)9597295. - Message d'origine De : elyakhlifi mustapha [EMAIL PROTECTED] À : R-help@stat.math.ethz.ch Envoyé le : Mercredi, 30 Mai 2007, 8h37mn 36s Objet : [R] functions without arguments hello, I wanna know if it's possible to write functions without argument in order to understand that I write down an example f - function(){ for (i in 1:length(C[[1]]) { print(10*C[[1]][i]) } } I know that there are errors in this syntax _ [[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 and provide commented, minimal, self-contained, reproducible code. ls.. et vos réactions ! [[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 and provide commented, minimal, self-contained, reproducible code.
[R] [ANN] Static and dynamic graphics course, July 2007, Salt Lake City
We're pleased to announce a one day course covering static and dynamic graphics using R, ggplot and GGobi. The course will be held just before the JSM, on Saturday, 28 July 2007, in Salt Lake City. The course will be presented by Dianne Cook and Hadley Wickham. In the course you will learn: * How to build presentation quality static graphics using the R package, ggplot. We will cover plot creation and modification, and discuss the grammar which underlies the package. * How to explore your data with direct manipulation/dynamic graphics using GGobi and rggobi. You'll learn the general toolbox, as well specific approaches for dealing with missing data, supervised classification, cluster analysis and multivariate longitudinal data analysis. Dianne Cook is a full professor at Iowa State University. She has been an active researcher in the field of interactive and dynamic graphics for 16 years, and regularly teaches information visualization, multivariate analysis and data mining. Hadley Wickham is a PhD student at Iowa State University. He won the John Chambers Award for statistical computing in 2006 for his work on ggplot. For more details, or to book your place, please see http://lookingatdata.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 and provide commented, minimal, self-contained, reproducible code.
[R] white test to check homoscedasticity of the residuals
Hi R-programmers, I can't find find the White test to check the homoscedasticity of the residuals from a linear model. Could you please help me with this? Thank you ! BC [[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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] search path question
Prof Brian Ripley wrote: You could do this via a search_file() connection wrapper, but there is a problem with ensuring connections get closed (which on.exit does here). I'm not sure exactly what you mean by a 'search_file() connection wrapper', but I have realised that its probably a better idea to write a function that checks a search path for a file and then returns that file: search_file = function(name,path=options()$scanpath,...){ for(p in path){ file=file.path(p,name) if(file.exists(file)){ return(file) } } return(name) } Then you can use that in any filename-using function: options(scanpath=c(/data1,/data2,/etc)) search_file(passwd) [1] /etc/passwd passwd = read.table(search_file(passwd),sep=:) record = scan(search_file(passwd),what='')[1] Barry __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] matrix in data.frame
Dear list! I have run into a problem that seems very simple but I can't find any solution to it (have searched the internet, help-files and An introduction to R etc without any luck). The problem is the following: I would like to create a data.frame with two components (columns), the first component being a matrix and the second component a vector. Whatever I have tried so far, I end up with a data.frame containing all the columns from the matrix plus the vector which is not what I am after. I have seen this kind of data.frame among R example datasets (oliveoil and yarn). I would greatly appreciate some help with this problem! Kind regards, Lina Hultin Rosenberg Karolinska Biomics Center Karolinska Institute Sweden __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Help me understand colours on linux
Hi Here is my sessionInfo(): Version 2.3.1 (2006-06-01) i686-redhat-linux-gnu attached base packages: [1] methods stats graphics grDevices utils datasets [7] base I have a function that is trying to draw rectangles using 136 different colours, and I get the following error: Error in rect(xstart, my.min, xstart + fcount[i, 2], my.max, col = fcolors[i], : Error: X11 cannot allocate additional graphics colours. Consider using X11 with colortype=pseudo.cube or gray. However, if I use pseudo.cube I don't get anywhere near enough distinct colours. I could use gray, but I would prefer colour. So, questions: 1) is there any set of options I can use which will actually let me create that many colours? 2) if not, how do I test if there is not, and implement gray instead of colours? This function and package works on windows, and it works with less colours. I guess I could try and trap the error, and if I can, go back and re-run the function with options(X11colortype=gray) but I'd prefer something more elegant... What I'm looking for is some test where I can say If you're going to fail, X11colortype='gray', else X11colortype='true'. Possible? Many thanks Mick The information contained in this message may be confidentia...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] functions without arguments
Sure it is possible, but it is a very bad and possibly dangerous idea, since you may not control what's passed to your function. Consider this example: x - 1:5 x [1] 1 2 3 4 5 foo - function(){mean(x)} foo() [1] 3 rm('x') foo() Error in mean(x) : object x not found It is much much beter to include argument(s) and to set them to some default value(s): foo - function(x=c(2,3,1,1,4,5,1)){mean(x)} foo() [1] 2.428571 foo(1:10) [1] 5.5 Petr elyakhlifi mustapha napsal(a): hello, I wanna know if it's possible to write functions without argument in order to understand that I write down an example f - function(){ for (i in 1:length(C[[1]]) { print(10*C[[1]][i]) } } I know that there are errors in this syntax _ [[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 and provide commented, minimal, self-contained, reproducible code. -- Petr Klasterecky Dept. of Probability and Statistics Charles University in Prague Czech Republic __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] matrix in data.frame
Have you thought of using a list? a - matrix(1:10, nrow=2) b - 1:5 x - list(a=a, b=b) x $a [,1] [,2] [,3] [,4] [,5] [1,]13579 [2,]2468 10 $b [1] 1 2 3 4 5 x$a [,1] [,2] [,3] [,4] [,5] [1,]13579 [2,]2468 10 x$b [1] 1 2 3 4 5 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Lina Hultin-Rosenberg Sent: 30 May 2007 10:26 To: r-help@stat.math.ethz.ch Subject: [R] matrix in data.frame Dear list! I have run into a problem that seems very simple but I can't find any solution to it (have searched the internet, help-files and An introduction to R etc without any luck). The problem is the following: I would like to create a data.frame with two components (columns), the first component being a matrix and the second component a vector. Whatever I have tried so far, I end up with a data.frame containing all the columns from the matrix plus the vector which is not what I am after. I have seen this kind of data.frame among R example datasets (oliveoil and yarn). I would greatly appreciate some help with this problem! Kind regards, Lina Hultin Rosenberg Karolinska Biomics Center Karolinska Institute Sweden __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] matrix in data.frame
You need to use I() or something similar. E.g. A - matrix(1:6, 2,3) data.frame(x=1:2, I(A)) X - data.frame(x=1:2) X$A - A both insert A as a single column. On Wed, 30 May 2007, Lina Hultin-Rosenberg wrote: Dear list! I have run into a problem that seems very simple but I can't find any solution to it (have searched the internet, help-files and An introduction to R etc without any luck). The problem is the following: I would like to create a data.frame with two components (columns), the first component being a matrix and the second component a vector. Whatever I have tried so far, I end up with a data.frame containing all the columns from the matrix plus the vector which is not what I am after. I have seen this kind of data.frame among R example datasets (oliveoil and yarn). I would greatly appreciate some help with this problem! Kind regards, Lina Hultin Rosenberg Karolinska Biomics Center Karolinska Institute Sweden -- 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] comparing fit of cubic spline
On Monday 21 May 2007 16:17, Thomas Reed wrote: I want to compare the fit of a quadratic model to continuous data, with that of a cubic spline fit. Is there a way of computing AIC from for e.g. a GAM with a smoothing spine, and comparing this to AIC from a quadratic model? library(mgcv) AIC(gam(y~s(x))) ## a `generalized aic' since effective df used AIC(gam(y~x+I(x^2))) Cheers ** Tom Reed PhD Student Institute of Evolutionary Biology 102 Ashworth Laboratories Kings Buildings University of Edinburgh Edinburgh EH9 3JT [EMAIL PROTECTED] Tel. 00 44 (0)131 6505462 http://homepages.ed.ac.uk/ecunning/tom.html http://homepages.ed.ac.uk/loeske/tom.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 and provide commented, minimal, self-contained, reproducible code. -- Simon Wood, Mathematical Sciences, University of Bath, Bath, BA2 7AY UK +44 1225 386603 www.maths.bath.ac.uk/~sw283 __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] matrix in data.frame
Thank you so much for your help, it worked of course! Best regards, Lina Hultin-Rosenberg -Ursprungligt meddelande- Från: Prof Brian Ripley [mailto:[EMAIL PROTECTED] Skickat: den 30 maj 2007 12:44 Till: Lina Hultin-Rosenberg Kopia: r-help@stat.math.ethz.ch Ämne: Re: [R] matrix in data.frame You need to use I() or something similar. E.g. A - matrix(1:6, 2,3) data.frame(x=1:2, I(A)) X - data.frame(x=1:2) X$A - A both insert A as a single column. On Wed, 30 May 2007, Lina Hultin-Rosenberg wrote: Dear list! I have run into a problem that seems very simple but I can't find any solution to it (have searched the internet, help-files and An introduction to R etc without any luck). The problem is the following: I would like to create a data.frame with two components (columns), the first component being a matrix and the second component a vector. Whatever I have tried so far, I end up with a data.frame containing all the columns from the matrix plus the vector which is not what I am after. I have seen this kind of data.frame among R example datasets (oliveoil and yarn). I would greatly appreciate some help with this problem! Kind regards, Lina Hultin Rosenberg Karolinska Biomics Center Karolinska Institute Sweden -- 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] search path question
I meant essentially the same thing but returning file(fixed_up_name). You can vectorize this BTW: search_file - function(name, path=getOption(scanpath)) { fp - c(name, file.path(path, name)) # better to use direct name first fp - fp[file.exists(fp)] if(length(fp)) file(fp[1]) else stop(name, not found) } Shortly things like read.table(search_file(passwd)) will close the unnamed connection for you. On Wed, 30 May 2007, Barry Rowlingson wrote: Prof Brian Ripley wrote: You could do this via a search_file() connection wrapper, but there is a problem with ensuring connections get closed (which on.exit does here). I'm not sure exactly what you mean by a 'search_file() connection wrapper', but I have realised that its probably a better idea to write a function that checks a search path for a file and then returns that file: search_file = function(name,path=options()$scanpath,...){ for(p in path){ file=file.path(p,name) if(file.exists(file)){ return(file) } } return(name) } Then you can use that in any filename-using function: options(scanpath=c(/data1,/data2,/etc)) search_file(passwd) [1] /etc/passwd passwd = read.table(search_file(passwd),sep=:) record = scan(search_file(passwd),what='')[1] Barry __ 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 and provide commented, minimal, self-contained, reproducible code. -- 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] How to analyse simple study: Placebo-controlled (2 groups) repeated measurements (ANOVA, ANCOA???)
Dear Frank Harrell, many thanks for your answers!!! I have downloaded your Harrell_notes.pdf and I think it would be best to have a look in some books you mentioned in the bibliographie. Can you recommend one especially? TO ANYBODY: Has anybody examples for similar data sets (as below) with R code? This would be a great help! (My impression of the example of Frank Harrell at the end of Harrell_notes.pdf is that it is more complicated and therefore more difficult to understand than the simplier data set I want to analyse.) Thanks!!! Karl - Ursprüngliche Mail Von: Frank E Harrell Jr [EMAIL PROTECTED] An: Karl Knoblick [EMAIL PROTECTED] CC: r-help@stat.math.ethz.ch Gesendet: Donnerstag, den 17. Mai 2007, 14:29:08 Uhr Betreff: Re: [R] How to analyse simple study: Placebo-controlled (2 groups) repeated measurements (ANOVA, ANCOA???) Karl Knoblick wrote: Hallo! I have two groups (placebo/verum), every subject is measured at 5 times, the first time t0 is the baseline measurement, t1 to t4 are the measurements after applying the medication (placebo or verum). The question is, if there is a significant difference in the two groups and how large the differnce is (95% confidence intervals). Let me give sample data # Data ID-factor(rep(1:50,each=5)) # 50 subjects GROUP-factor(c(rep(Verum, 115), rep(Placebo, 135))) TIME-factor(rep(paste(t,0:4,sep=), 50)) set.seed(1234) Y-rnorm(250) # to have an effect: Y[GROUP==Verum TIME==t1]-Y[GROUP==Verum TIME==t1] + 0.6 Y[GROUP==Verum TIME==t2]-Y[GROUP==Verum TIME==t2] + 0.3 Y[GROUP==Verum TIME==t3]-Y[GROUP==Verum TIME==t3] + 0.9 Y[GROUP==Verum TIME==t4]-Y[GROUP==Verum TIME==t4] + 0.9 DF-data.frame(Y, ID, GROUP, TIME) I have heard of different ways to analyse the data 1) Comparing the endpoint t4 between the groups (t-test), ignoring baseline Don't even consider this 2) Comparing the difference t4 minus t0 between the two groups (t-test) This is not optimal 3) Comparing the endpoint t4 with t0 as a covariate between the groups (ANOVA - how can this model be calculated in R?) Using t0 as a covariate is the way to go. A question is whether to just use t4. Generally this is not optimum. 4) Taking a summary score (im not sure but this may be a suggestion of Altman) istead of t4 5) ANOVA (repeated measurements) times t0 to t5, group placebo/verum), subject as random factor - interested in interaction times*groups (How to do this in R?) 6) as 5) but times t1 to t5, ignoring baseline (How to do this in R?) 7) as 6) but additional covariate baseline t0 (How to do this in R?) What will be best? - (Advantages / disadvantages?) How to analyse these models in R with nested and random effects and possible covariate(ID, group - at least I think so) and random parameter ID)? Or is there a more simple possibility? It's not obvious that random effects are needed if you take the correlation into account in a good way. Generalized least squares using for example an AR1 correlation structure (and there are many others) is something I often prefer. A detailed case study with R code (similar to your situation) is in http://biostat.mc.vanderbilt.edu/FrankHarrellGLS . This includes details about why t0 is best to consider as a covariate. One reason is that the t0 effect may not be linear. If you want to focus on t4 it is easy to specify a contrast (after fitting is completed) that tests t4. If time is continuous this contrast would involve predicted values at the 4th time, otherwise testing single parameters. Frank Harrell Perhaps somebody can recommend a book or weblink where these different strategies of analysing are discussed - preferable with examples with raw data which I can recalculate. And if there is the R syntax includede - this would be best! Any help will be appreciate! Thanks! Karl -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ Yahoo! Clever: Stellen Sie Fragen und finden Sie Antworten. Teilen Sie Ihr Wissen. www.yahoo.de/clever __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] missing values
hello all i would like to perform multiple imputation using the norm library. but i seem to get the following error when i use the da.norm function. Error in as.double.default(list(V1 = c(0.058177827, 0.123076923, 0.138713745, : (list) object cannot be coerced to 'double' can anyone help? thanking you in advance Allan Clark Lecturer in Statistical Sciences Department University of Cape Town 7701 Rondebosch South Africa TEL (Office): +27-21-650-3228 FAX: +27-21-650-4773 http://web.uct.ac.za/depts/stats/aclark.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 and provide commented, minimal, self-contained, reproducible code.
[R] test to compare significant correlation increase
Hi! I am calculating correlation between two variables: 1. X versus Y 2. X versus Y(with a 3 steps lag) I would like to test if the correlation increase/decrease from 1 to 2 is significant or not. Is there any function in R to do this? any hints? Thanks for help :) David Riaño Center for Spatial Technologies and Remote Sensing (CSTARS) University of California 250-N, The Barn One Shields Avenue Davis, CA 95616-8527 USA 1-(517) 629-5499 http://www.cstars.ucdavis.edu/~driano/index.html http://www.cstars.ucdavis.edu __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] white test to check homoscedasticity of the residuals
On Wed, 30 May 2007, Benoit Chemineau wrote: Hi R-programmers, I can't find find the White test to check the homoscedasticity of the residuals from a linear model. Could you please help me with this? The package lmtest includes the function bptest() for performing Breusch-Pagan tests. White's test is a special case of this. For example, if you fit a linear regression fm - lm(y ~ x + z, data = foo) then you can carry out White's test via bptest(fm, ~ x * z + I(x^2) + I(z^2), data = foo) i.e., include all regressors and the squares/cross-products in the auxiliary regression. I haven't yet written a simple convenience interface for this... hth, Z Thank you ! BC [[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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Smoothing a path in 2D
Hello, I'm currently trying to find a method to interpolate or smooth data that represent a trajectory in space. For example, I have an ordered (=time) set of (x,y) tuples which constitute a path in a 2D space. Is there a way using R to interpolate between these points in a way similar to spline interpolation so that I get a smooth path in space? Greetings, Dieter -- Dieter Vanderelst [EMAIL PROTECTED] Department of Industrial Design Designed Intelligence [[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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Smoothing a path in 2D
You might have a look at the fda package of Ramsay on CRAN. url:www.econ.uiuc.edu/~rogerRoger Koenker email[EMAIL PROTECTED]Department of Economics vox: 217-333-4558University of Illinois fax: 217-244-6678Champaign, IL 61820 On May 30, 2007, at 9:42 AM, Dieter Vanderelst wrote: Hello, I'm currently trying to find a method to interpolate or smooth data that represent a trajectory in space. For example, I have an ordered (=time) set of (x,y) tuples which constitute a path in a 2D space. Is there a way using R to interpolate between these points in a way similar to spline interpolation so that I get a smooth path in space? Greetings, Dieter -- Dieter Vanderelst [EMAIL PROTECTED] Department of Industrial Design Designed Intelligence [[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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] codamenu() :Error in coda.options....
Dear all, I recently started having some problemd with the coda package. I have also deleted it and then installed the package again. The problem is that I cant access the codamenu. That is ; * library(coda) codamenu() Error in coda.options(default = TRUE) : cannot change value of locked binding for '.Coda.Options' Quitting codamenu ... Error in !coda.options(data.saved) : invalid argument type * Thank you, Regards, Luwis Diya [[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 and provide commented, minimal, self-contained, reproducible code.
[R] [R-pkgs] Revised Rcmdr.HH package
I posted a revised Rcmdr.HH_1.8-0 package on CRAN. The Rcmdr.HH package adds additional menu items to the Rcmdr package by John Fox. Our introductory course at Temple University includes several topics that were not addressed in the Rcmdr. This revision uses the new RcmdrPlugin technology that John announced a few days ago. Rich ___ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] missing values
When it says 'matrix' it means it, not 'data frame'. On Wed, 30 May 2007, Allan Clark wrote: hello all i would like to perform multiple imputation using the norm library. but i seem to get the following error when i use the da.norm function. Error in as.double.default(list(V1 = c(0.058177827, 0.123076923, 0.138713745, : (list) object cannot be coerced to 'double' can anyone help? thanking you in advance Allan Clark Lecturer in Statistical Sciences Department University of Cape Town 7701 Rondebosch South Africa TEL (Office): +27-21-650-3228 FAX: +27-21-650-4773 http://web.uct.ac.za/depts/stats/aclark.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 and provide commented, minimal, self-contained, reproducible code. -- 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Smoothing a path in 2D
?KalmanLike Clint BowmanINTERNET: [EMAIL PROTECTED] Air Dispersion Modeler INTERNET: [EMAIL PROTECTED] Air Quality Program VOICE: (360) 407-6815 Department of Ecology FAX:(360) 407-7534 USPS: PO Box 47600, Olympia, WA 98504-7600 Parcels:300 Desmond Drive, Lacey, WA 98503-1274 On Wed, 30 May 2007, Dieter Vanderelst wrote: Hello, I'm currently trying to find a method to interpolate or smooth data that represent a trajectory in space. For example, I have an ordered (=time) set of (x,y) tuples which constitute a path in a 2D space. Is there a way using R to interpolate between these points in a way similar to spline interpolation so that I get a smooth path in space? Greetings, Dieter -- Dieter Vanderelst [EMAIL PROTECTED] Department of Industrial Design Designed Intelligence [[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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] separate y-limits in xYplot panels
Hello, I would like to get the scales of y-axes dependent only on the data points in a particular panel. Have attached a test example below. When using 'relation=free', it does not make the scales 'free', however when using 'relation=sliced', I get a warning Explicitly specified limits ignored in: limitsFromLimitlist(have.lim = have.ylim, lim = ylim, relation = y.relation, (although in this particular case, I get the desired result, but in my real data, I do not get the free y-scale for each panel). Can you please let me know what should be correct syntax? Thanks. -Nitin library(Hmisc) test1 - data.frame( y = c(rnorm(33), rnorm(33, mean=10), rnorm(34, mean=100)), x = 1:100, f = factor(c(rep(a, 33), rep(b, 33), rep(c, 34))), g = factor(sample(LETTERS[1:2], size=100, replace=TRUE)) ) CI - rnorm(100) lb - test1$y - CI ub - test1$y + CI xYplot(Cbind(y, lb,ub )~x|f, groups=g, scales = list(relation=free), ## Changing it to sliced gives warning data=test1) Don't pick lemons. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] How to check for existence url from within a function?
Duncan == Duncan Murdoch [EMAIL PROTECTED] on Sat, 26 May 2007 08:02:11 -0400 writes: Duncan On 26/05/2007 7:13 AM, Heinz Tuechler wrote: Dear All, To check if an url exists, I can use try(). This works, as I expected, if I do it directly, as in the first part of the following example, but I could [..] Duncan .Last.value isn't set until your function returns. You should write this as Duncan con.url - try(url(url.string, open='rb')) Duncan try.error - inherits(con.url, try-error) Duncan Notice that I used inherits, rather than testing for equality. It's Duncan documented that the result of try() will be of class 'try-error' if an Duncan error occurs, but there may be circumstances (in the future?) where Duncan different types of errors are signalled by using a more complicated class. There's an additional reason for inherits(.,.) and hence against class(foo) == bar : Whenever try() does not catch an error, since there was none, i.e., the result of try(foobar(..)) is just foobar(..), foobar(..) may well return an object with an (S3) class vector of length 1. In those cases, the equality test returns a logical vector, typically c(FALSE, FALSE) and using that in if(.) gives at least a warning. 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] http proxies: setting and unsetting
There is also a free program setenv.exe which is more powerful than setx.exe. You mentioned that deleting http_proxy from your environment through the OS would fix your problem. setenv.exe can both set and delete environment variables and you can specify user, system, etc. See description and link here: http://www.jsifaq.com/SF/Tips/Tip.aspx?id=4928 Grothendieck [EMAIL PROTECTED] wrote: One other point. If you find you need to set a system or user environment variable then microsoft has a free tool called setx.exe that you can find here: http://support.microsoft.com/kb/927229 You can do this from within R using system(). On 5/30/07, Gabor Grothendieck [EMAIL PROTECTED] wrote: On 5/29/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: OK, I think I get that... do you know which namespace the Sys.setenv() function affects? Do you know if there are functions in R that can alter the user/system/process environment variables? Use the R Sys.getenv() command to get the process environment variables. To get user and system environment variables, from the Desktop right click on My Computer and choose Properties. Then choose the Advanced tab and click on the Environment Variables button near the bottom of the window that appears. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] http proxies: setting and unsetting
Thanks... I'll check the documentation to see which file on Windows I need to alter to issue one of those commands. matt -Original Message- From: Prof Brian Ripley [mailto:[EMAIL PROTECTED] Sent: Wednesday, May 30, 2007 12:39 AM To: Gabor Grothendieck Cc: Pettis, Matthew (Thomson); r-help@stat.math.ethz.ch Subject: Re: [R] http proxies: setting and unsetting That was misleading advice. R is a C program and accesses the environment via the C calls getenv and (on Windows) putenv. This is not Windows scripting (Grothendieck's earlier reference): the C runtime maintains only one environment block. I've re-checked, and suspect the problem is in the following comment in ?download.file These environment variables must be set before the download code is first used: they cannot be altered later by calling 'Sys.setenv'. If I have http_proxy set in the environment, the proxy is not used in any of the following cases: - Calling R from a shortcut with http_proxy= on the target (see the rw-FAQ). - Calling R from a shortcut with no_proxy=* on the target. - Using Sys.unsetenv(http_proxy) right at the start of the R session. - Using Sys.setenv(no_proxy=*) right at the start of the R session. If you set options(internet.info=0) you will see exactly what is tried. Another possibility is that R is being used with --internet2, in which case none of this applies, and the simple answer is not to use --internet2 at home. On Tue, 29 May 2007, Gabor Grothendieck wrote: You can have 4 different http_proxy environment variables and if you set one type but try to unset a different type then that will have no effect on the one you originally set. For example, if you originally set it as a system or user environment variable and then try to unset the process environment variable of the same name then that will have no effect on the system or user environment variable. On 5/29/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hi Gabor, Thanks for the reply and link. I took a look at the link -- one thing I don't understand is why if I delete the 'http_proxy' variable via the cmd shell (or equivalent OS dialog box), why I can get R to ignore the proxy, but using Sys.setenv(http_proxy=) won't do that for me (at least for the scope of the session). If there were other variables affecting it, I would think my deleting 'http_proxy' in the OS would also have no effect -- yet it does. Any ideas? Thanks again, Matt -Original Message- From: Gabor Grothendieck [mailto:[EMAIL PROTECTED] Sent: Tue 5/29/2007 9:49 PM To: Pettis, Matthew (Thomson) Cc: r-help@stat.math.ethz.ch Subject: Re: [R] http proxies: setting and unsetting Note that Windows XP has 4 types of environment variables and I suspect that the problem stems from not taking that into account: http://www.microsoft.com/technet/scriptcenter/guide/sas_wsh_kmmj.mspx ?mfr=true On 5/29/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hi, I am trying to use R at work and at home on the same computer. At work, I have a proxy, and at home, I do not. I have, for work, a User environment variable http_proxy which I set in the OS (Windows XP Pro). When I am at work, and I try to retrieve data from the web with 'read.csv', things work just fine. I assume it knows how to use the proxy. The trouble is when I am at home and have no proxy, R still tries to use my work proxy. I have tried the following: Sys.setenv(http_proxy=) Sys.setenv(no_proxy=TRUE) Sys.setenv(no_proxy=1) none of which seems to work. Whenever I try to use read.csv, it tells me that it cannot find my work proxy, which I am trying to tell R to ignore. I can solve this problem by removing the http_proxy environment variable binding in the OS when at home, but that is a pain, because then I have to reset it when I go back into work. Is there a way to tell R within a session to ignore the proxy? If so, what am I doing wrong? thanks, matt __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code. -- 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
Re: [R] separate y-limits in xYplot panels
Nitin Jain said the following on 5/30/2007 8:12 AM: Hello, I would like to get the scales of y-axes dependent only on the data points in a particular panel. Have attached a test example below. When using 'relation=free', it does not make the scales 'free', however when using 'relation=sliced', I get a warning Explicitly specified limits ignored in: limitsFromLimitlist(have.lim = have.ylim, lim = ylim, relation = y.relation, (although in this particular case, I get the desired result, but in my real data, I do not get the free y-scale for each panel). Can you please let me know what should be correct syntax? Thanks. -Nitin library(Hmisc) test1 - data.frame( y = c(rnorm(33), rnorm(33, mean=10), rnorm(34, mean=100)), x = 1:100, f = factor(c(rep(a, 33), rep(b, 33), rep(c, 34))), g = factor(sample(LETTERS[1:2], size=100, replace=TRUE)) ) CI - rnorm(100) lb - test1$y - CI ub - test1$y + CI xYplot(Cbind(y, lb,ub )~x|f, groups=g, scales = list(relation=free), ## Changing it to sliced gives warning data=test1) You want scales = list(y = list(relation = free)) 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] http proxies: setting and unsetting
Thanks... I'll give it a whirl... matt -Original Message- From: Gabor Grothendieck [mailto:[EMAIL PROTECTED] Sent: Wednesday, May 30, 2007 10:22 AM To: Pettis, Matthew (Thomson) Cc: r-help@stat.math.ethz.ch Subject: Re: [R] http proxies: setting and unsetting There is also a free program setenv.exe which is more powerful than setx.exe. You mentioned that deleting http_proxy from your environment through the OS would fix your problem. setenv.exe can both set and delete environment variables and you can specify user, system, etc. See description and link here: http://www.jsifaq.com/SF/Tips/Tip.aspx?id=4928 Grothendieck [EMAIL PROTECTED] wrote: One other point. If you find you need to set a system or user environment variable then microsoft has a free tool called setx.exe that you can find here: http://support.microsoft.com/kb/927229 You can do this from within R using system(). On 5/30/07, Gabor Grothendieck [EMAIL PROTECTED] wrote: On 5/29/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: OK, I think I get that... do you know which namespace the Sys.setenv() function affects? Do you know if there are functions in R that can alter the user/system/process environment variables? Use the R Sys.getenv() command to get the process environment variables. To get user and system environment variables, from the Desktop right click on My Computer and choose Properties. Then choose the Advanced tab and click on the Environment Variables button near the bottom of the window that appears. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] learning lattice graphics
On 2007-05-27, Sundar Dorai-Raj [EMAIL PROTECTED] wrote: I would also suggest Paul Murrell's book R Graphics. http://www.amazon.com/Graphics-Computer-Science-Data-Analysis/dp/158488486X/ --sundar Thanks all! I have the relevant chapters from the Murrell book now (which were available as samples), and between that, the bell labs site, and the help pages I think I'll be ok! Tyler __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] manova permutations and pair-wise contrasts
Hi, I have a function for doing permutation tests for Manovas, written with the help of folks here on the list. It seems to work ok, and I've found that there is indeed a significant difference among groups in my analysis. I want to follow up on this by testing for which pairs of groups are significantly different. Reasoning that since Wilks Lambda is a generalization of the Hotelling T test, it should be appropriate for testing difference between two groups as well as between many groups, I used the same function. This appears to work, with one exception. For one pair of groups the residuals have rank 14 15. Reading through the archives I see that this problem comes up with permutation tests on occassion. My question is, does this problem indicate a mistaken assumption or faulty code on my part, and is there anyway to address it, either with a different approach or a change to my function? The actual function I use is: wilks.perm - function(data.mat, cat.vect, n=999){ res.vect - numeric(n+1) for (i in 1:n){ wilks - summary(manova(data.mat ~ sample(cat.vect)), test=Wilks)$stats[1,2] res.vect[i] - wilks } res.vect[n+1] - obs - summary(manova(data.mat~cat.vect), test=Wilks)$stats[1,2] return(sum(res.vect = res.vect[n+1])/(n+1)) } Thanks! Tyler __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] separate y-limits in xYplot panels
You could consider doing this directly with lattice. See: demo(intervals) On 5/30/07, Nitin Jain [EMAIL PROTECTED] wrote: Hello, I would like to get the scales of y-axes dependent only on the data points in a particular panel. Have attached a test example below. When using 'relation=free', it does not make the scales 'free', however when using 'relation=sliced', I get a warning Explicitly specified limits ignored in: limitsFromLimitlist(have.lim = have.ylim, lim = ylim, relation = y.relation, (although in this particular case, I get the desired result, but in my real data, I do not get the free y-scale for each panel). Can you please let me know what should be correct syntax? Thanks. -Nitin library(Hmisc) test1 - data.frame( y = c(rnorm(33), rnorm(33, mean=10), rnorm(34, mean=100)), x = 1:100, f = factor(c(rep(a, 33), rep(b, 33), rep(c, 34))), g = factor(sample(LETTERS[1:2], size=100, replace=TRUE)) ) CI - rnorm(100) lb - test1$y - CI ub - test1$y + CI xYplot(Cbind(y, lb,ub )~x|f, groups=g, scales = list(relation=free), ## Changing it to sliced gives warning data=test1) Don't pick lemons. __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] codamenu() :Error in coda.options....
Please report bugs in packages to the package maintainer (CCing) including version information and platform such as for me in this case: Same for me, R-2.5.0, Windows XP, coda 0.11-1. Uwe Ligges Luwis Diya wrote: Dear all, I recently started having some problemd with the coda package. I have also deleted it and then installed the package again. The problem is that I cant access the codamenu. That is ; * library(coda) codamenu() Error in coda.options(default = TRUE) : cannot change value of locked binding for '.Coda.Options' Quitting codamenu ... Error in !coda.options(data.saved) : invalid argument type * Thank you, Regards, Luwis Diya [[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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] test to compare significant correlation increase
Just a guess (please correct if I'm way off on this), but maybe you could look at the difference in r betwen 1 2 and see if the confidence interval (http://davidmlane.com/hyperstat/B8544.html) for this value given your sample size includes 0. On 30-May-07, at 11:24 AM, David Riano wrote: Hi! I am calculating correlation between two variables: 1. X versus Y 2. X versus Y(with a 3 steps lag) I would like to test if the correlation increase/decrease from 1 to 2 is significant or not. Is there any function in R to do this? any hints? Thanks for help :) David Riaño Center for Spatial Technologies and Remote Sensing (CSTARS) University of California 250-N, The Barn One Shields Avenue Davis, CA 95616-8527 USA 1-(517) 629-5499 http://www.cstars.ucdavis.edu/~driano/index.html http://www.cstars.ucdavis.edu __ 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 and provide commented, minimal, self-contained, reproducible code. -- Mike Lawrence Graduate Student, Department of Psychology, Dalhousie University Website: http://myweb.dal.ca/mc973993 Public calendar: http://icalx.com/public/informavore/Public The road to wisdom? Well, it's plain and simple to express: Err and err and err again, but less and less and less. - Piet Hein __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] separate y-limits in xYplot panels
Thanks Sundar and Gabor for your prompt help. Sundar - Even after changing the line to scales = list(y = list(relation = free)), I do not get the free scales in my test code (below), and relation =sliced still gives warning. Gabor - I'll try using code as demo(intervals) suggests. Best, Nitin Nitin Jain said the following on 5/30/2007 8:12 AM: Hello, I would like to get the scales of y-axes dependent only on the data points in a particular panel. Have attached a test example below. When using 'relation=free', it does not make the scales 'free', however when using 'relation=sliced', I get a warning Explicitly specified limits ignored in: limitsFromLimitlist(have.lim = have.ylim, lim = ylim, relation = y.relation, (although in this particular case, I get the desired result, but in my real data, I do not get the free y-scale for each panel). Can you please let me know what should be correct syntax? Thanks. -Nitin library(Hmisc) test1 - data.frame( y = c(rnorm(33), rnorm(33, mean=10), rnorm(34, mean=100)), x = 1:100, f = factor(c(rep(a, 33), rep(b, 33), rep(c, 34))), g = factor(sample(LETTERS[1:2], size=100, replace=TRUE)) ) CI - rnorm(100) lb - test1$y - CI ub - test1$y + CI xYplot(Cbind(y, lb,ub )~x|f, groups=g, scales = list(relation=free), ## Changing it to sliced gives warning data=test1) You want HTH, --sundar Choose the right car based on your needs. Check out Yahoo! Autos new Car Finder tool. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] test to compare significant correlation increase
In response to the original message, see http://finzi.psych.upenn.edu/R/Rhelp02a/archive/85753.html which describes two papers about comparing dependent correlations, which I think is a relevant question. (Try Google too. I didn't.) There is also a paper by Steiger in Psychological Bulletin, 1980, which I implemented in the following R code. (Not tested since I translated it from a BASIC script written in 1981.) r12 - as.numeric(readline(Difference r12 and r13, given r23. Input r12: )) r13 - as.numeric(readline(r13: )) r23 - as.numeric(readline(r23: )) N - as.numeric(readline(Number of Ss: )) rd - 1-r12*r12-r13*r13-r23*r23+2*r12*r13*r23 t2 - (r12-r13)*sqrt((N-1)*(1+r23)/(2*rd*(N-1)/(N-3)+((r12+r13)/2)^2*(1-r23)^3)) print(paste(t(,N-3,)=,t2, p=,pt(t2,N-3), one tailed,sep=)) print(Steiger, J.H. (1980). Tests for comparing elements of a) print(correlation matrix. Psychological Bulletin, 87, 245-251.) On 30-May-07, at 11:24 AM, David Riano wrote: Hi! I am calculating correlation between two variables: 1. X versus Y 2. X versus Y(with a 3 steps lag) I would like to test if the correlation increase/decrease from 1 to 2 is significant or not. Is there any function in R to do this? any hints? Thanks for help :) David Ria�o Center for Spatial Technologies and Remote Sensing (CSTARS) University of California 250-N, The Barn One Shields Avenue Davis, CA 95616-8527 USA 1-(517) 629-5499 http://www.cstars.ucdavis.edu/~driano/index.html http://www.cstars.ucdavis.edu __ 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 and provide commented, minimal, self-contained, reproducible code. -- Mike Lawrence Graduate Student, Department of Psychology, Dalhousie University Website: http://myweb.dal.ca/mc973993 Public calendar: http://icalx.com/public/informavore/Public The road to wisdom? Well, it's plain and simple to express: Err and err and err again, but less and less and less. - Piet Hein __ 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 and provide commented, minimal, self-contained, reproducible code. -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page: http://www.sas.upenn.edu/~baron __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Empirical Complementary CDF
Hi all, I'd like to plot an empirical LLCD (log-log-CCDF, where CCDF == 1-CDF). It seems ecdf() and plot(ecdf()) can't do it. I'd like to reuse some graphics features of ECDF plot. The fastest way I've found is to copypaste the ecdf code and writing this function: --- [R-code] --- eccdf - function(x) { x - sort(x) n - length(x) if (n 1) stop('x' must have 1 or more non-missing values) vals - sort(unique(x)) rval - approxfun(vals, 1-cumsum(tabulate(match(x, vals)))/n, #[CHANGED] method = constant, yleft = 1, yright = 0, f = 0, ties = ordered) class(rval) - c(eccdf, stepfun, class(rval)) #[CHANGED] attr(rval, call) - sys.call() rval } --- [/R-code] --- and the rewriting the plot.ecdf function: --- [R code] --- plot.eccdf - function (x, ..., ylab = 1-Fn(x), verticals = FALSE, col.01line = gray70) #CHANGED { plot.stepfun(x, ..., ylab = ylab, verticals = verticals) abline(h = c(0, 1), col = col.01line, lty = 2) } --- [/R-code] --- So, is there a even faster way to get it? Thank you so much!! -- Marco __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] checking for viability of a GUI component
Dear All: The question: How do I check for existance of the GUI, instead of checking the variable of the GUI? I created a log window for several applications, they will check for the existance of log window, if it exists, append output to it, otherwise, create log window and insert to it. What I found out is that if I close the X window of the GUI, the variable logwin still exists, confuse arises to other components that are supposed to use the GUI. create.log.win - function(inputtext){ if(exists(is.null(logwin))) {return} else { logwin - tktoplevel() logtext - tktext(logwin, bg=white) tkwm.title(logwin, Log Window) loglabel - tklabel(logwin, text=Logging Analaysis Information) tkgrid(loglabel) tkgrid(logtext) if (is.matrix(inputtext)) { for (i in 1:nrow(inputtext)) tkinsert(logtext, end,paste(paste(inputtext[i,],collapse= ),\n,sep=)) } else tkinsert(logtext, end,paste(inputtext,\n,sep=)) export_button - tkbutton(logwin, text = Export Log, command=savelog) tkgrid(export_button) } } -- I understand I should bind a function to remove the varialbe with a close button, however, we can't stop users from closing window just by clicking on the upper right corner. How do I check for existance of the GUI, instead of the variable? Thanks __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Connecting to PostgreSQL/PostGIS from R (rgdal?)
Hello, I've been trying every now and then to find a cross operating system solution that would let me access PostgreSQL (and PostGIS) from R, or to access R from PostgreSQL. I know of pl/r, which accomplishes the latter, but has yet to be successfully ported to Windows. Similarly, I've tried to use Rdbi and DBI, but I haven't had luck with those on Windows either for connecting to PostgreSQL from R. Can anyone suggest a solution for this? It would seem that rgdal could also help me in this case. Unfortunately, the version of the GDAL library that is included in the rdgal binary available on CRAN (for windows) doesn't include the PostgreSQL driver for OGR (i.e., it's not listed by the ogrDrivers() function). I compiled rgdal on Windows myself using the GDAL library from FWTools-1.3.1, but I was unsuccessful at creating a proper binary package for R. I was only able to get it to work by substituting the rgdal.dll that was installed by CRAN with the one that I compiled that links against the GDAL library from FWTools. Even though it works (at first glance with ogrInfo(), and readOGR()), I still get a warning message when I load the libary: DLL attempted to change FPU control word from 8001f to 9001f. So my question with respect to rgdal is a) is it likely that an rgdal package is going to be released in the future with the PostgreSQL driver included in GDAL/OGR, or b) are there any suggestions/instructions that might get me through the compilation and packaging process for rgdal with better success? Thanks in advance for any help, Mike __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] riv-package, how to deal with missing values?
Hello everyone! I want to calculate an instrumental variable estimator using the riv-package: riv(Y,Xex,Xend,W,method=classical) and my problem is that I either get an error because missing values are not allowed in the matrices or when I exclude NA's from the matrices it says that matrices have to be at the same length. Is there a special option how to deal with missing values when using the riv-package? Thanks to everyone how might be able to help! Jan Störger Mannheim 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 and provide commented, minimal, self-contained, reproducible code.
[R] opinions please: text editors and reporting/Sweave?
dear all - I currently use Tinn-R as my text editor to work with code that I submit to R, with some output dumped to text files, some images dumped to pdf. (system: Windows 2K and XP, R 2.4.1 and R 2.5). We are using R for overnight runs to create large output data files for GIS, but then I need simple output reports for analysis results for each separate data set. Thus, I create many reports of the same style, but just based on different input data. I am recognizing that I need a better reporting system, so that I can create clean reports for each separate R run. This obviously means using Sweave and some implementation of LaTex, both of which are new to me. I've installed MikTex and successfully completed a demo or two for creating pdfs from raw LaTeX. It appears that if I want to ease my entry into the world of LaTeX, I might need to switch editors to something like Emacs (I read somewhere that Emacs helps with the TeX markup?). After quite a while wallowing at the Emacs site, I am finding that ESS is well integrated with R and might be the way to go. Aaaagh... I'm in way over my head! My questions: What, in your opinion, is the simplest way to integrate text and graphics reports into a single report such as a pdf file. If the answer to this is LaTeX and Sweave, is it difficult to use a text editor such as Tinn-R or would you strongly recommend I leave behind Tinn and move over to an editor that has more LaTeX help? In reading over Friedrich Leisch's Sweave User Manual (v 1.6.0) I am beginning to think I can do everything I need with my simple editor. Before spending many hours going down that path, I thought it prudent to ask the R community. It is likely I am misunderstanding some of the process here and any clarifications are welcome. Thank you in advance for any thoughts. Tim Howard __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] [R-sig-Geo] Connecting to PostgreSQL/PostGIS from R (rgdal?)
On Wed, 30 May 2007, Mike Leahy wrote: Hello, I've been trying every now and then to find a cross operating system solution that would let me access PostgreSQL (and PostGIS) from R, or to access R from PostgreSQL. I know of pl/r, which accomplishes the latter, but has yet to be successfully ported to Windows. Similarly, I've tried to use Rdbi and DBI, but I haven't had luck with those on Windows either for connecting to PostgreSQL from R. Can anyone suggest a solution for this? It would seem that rgdal could also help me in this case. Unfortunately, the version of the GDAL library that is included in the rdgal binary available on CRAN (for windows) doesn't include the PostgreSQL driver for OGR (i.e., it's not listed by the ogrDrivers() function). I compiled rgdal on Windows myself using the GDAL library from FWTools-1.3.1, but I was unsuccessful at creating a proper binary package for R. I was only able to get it to work by substituting the rgdal.dll that was installed by CRAN with the one that I compiled that links against the GDAL library from FWTools. Even though it works (at first glance with ogrInfo(), and readOGR()), I still get a warning message when I load the libary: DLL attempted to change FPU control word from 8001f to 9001f. So my question with respect to rgdal is a) is it likely that an rgdal package is going to be released in the future with the PostgreSQL driver included in GDAL/OGR, or b) are there any suggestions/instructions that might get me through the compilation and packaging process for rgdal with better success? The warning is harmless - R is just reporting that it has stopped the dynamically linked libraries resetting a flag that they should not change while R is running. If you followed the notes in README.windows in rgdal, you ought to be OK. There are no plans to provide more Windows binary drivers than those present now, because the others involve further external dependencies, which most users would not welcome. So please try to do an ogrinfo at the command line in Windows using FWTools to your PostGIS data, and then the equivalent within R with your locally built rgdal, and see how it goes. Even on Linux, getting all the components lined up isn't easy, according to people who have tried, but can be done if you need to do it. Hope this helps, Roger Thanks in advance for any help, Mike ___ R-sig-Geo mailing list [EMAIL PROTECTED] https://stat.ethz.ch/mailman/listinfo/r-sig-geo -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] opinions please: text editors and reporting/Sweave?
Have you tried R2HTML, or is HTML not what you're looking for? -Original Message- From: [EMAIL PROTECTED] on behalf of Tim Howard Sent: Wed 30/05/2007 9:43 PM To: r-help@stat.math.ethz.ch Subject: [R] opinions please: text editors and reporting/Sweave? dear all - I currently use Tinn-R as my text editor to work with code that I submit to R, with some output dumped to text files, some images dumped to pdf. (system: Windows 2K and XP, R 2.4.1 and R 2.5). We are using R for overnight runs to create large output data files for GIS, but then I need simple output reports for analysis results for each separate data set. Thus, I create many reports of the same style, but just based on different input data. I am recognizing that I need a better reporting system, so that I can create clean reports for each separate R run. This obviously means using Sweave and some implementation of LaTex, both of which are new to me. I've installed MikTex and successfully completed a demo or two for creating pdfs from raw LaTeX. It appears that if I want to ease my entry into the world of LaTeX, I might need to switch editors to something like Emacs (I read somewhere that Emacs helps with the TeX markup?). After quite a while wallowing at the Emacs site, I am finding that ESS is well integrated with R and might be the way to go. Aaaagh... I'm in way over my head! My questions: What, in your opinion, is the simplest way to integrate text and graphics reports into a single report such as a pdf file. If the answer to this is LaTeX and Sweave, is it difficult to use a text editor such as Tinn-R or would you strongly recommend I leave behind Tinn and move over to an editor that has more LaTeX help? In reading over Friedrich Leisch's Sweave User Manual (v 1.6.0) I am beginning to think I can do everything I need with my simple editor. Before spending many hours going down that path, I thought it prudent to ask the R community. It is likely I am misunderstanding some of the process here and any clarifications are welcome. Thank you in advance for any thoughts. Tim Howard __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] opinions please: text editors and reporting/Sweave?
Tim, First, I personnally am a big fan of LaTeX, Emacs, and ESS and I think that in the long run you would benefit from learning all of them (probably start with Emacs, then ESS, then LaTeX once you already have a knowledge of Emacs and how it can help). Since you asked about the simplest way to go, you may want to look at the odfWeave package. This gives you the power of Sweave, but without having to learn LaTeX. It works with documents in openoffice (a free office suite similar to and mostly compatible with microsoft office (word)). Using this you can create your template using openoffice writer (or MS word, then convert using openoffice), run it through R/odfWeave, and have the result as another openoffice document that can then be converted to MS word or pdf. Personally, if I am doing something for myself, or in which the output format does not matter, then I use Sweave with LaTeX (using Emacs and ESS). But, often my results need to be sent to a client that will cut and past my results into an MS word document or power point presentation. Then I find it easier to use openoffice and odfWeave and have the end result be an MS word document that can be e-mailed to the client for them to mangle in ways they feel the need to. (there is also an HTML driver for Sweave I nthe the R2HTML package where you can have the template file resemble html and the final output is in html) Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tim Howard Sent: Wednesday, May 30, 2007 2:43 PM To: r-help@stat.math.ethz.ch Subject: [R] opinions please: text editors and reporting/Sweave? dear all - I currently use Tinn-R as my text editor to work with code that I submit to R, with some output dumped to text files, some images dumped to pdf. (system: Windows 2K and XP, R 2.4.1 and R 2.5). We are using R for overnight runs to create large output data files for GIS, but then I need simple output reports for analysis results for each separate data set. Thus, I create many reports of the same style, but just based on different input data. I am recognizing that I need a better reporting system, so that I can create clean reports for each separate R run. This obviously means using Sweave and some implementation of LaTex, both of which are new to me. I've installed MikTex and successfully completed a demo or two for creating pdfs from raw LaTeX. It appears that if I want to ease my entry into the world of LaTeX, I might need to switch editors to something like Emacs (I read somewhere that Emacs helps with the TeX markup?). After quite a while wallowing at the Emacs site, I am finding that ESS is well integrated with R and might be the way to go. Aaaagh... I'm in way over my head! My questions: What, in your opinion, is the simplest way to integrate text and graphics reports into a single report such as a pdf file. If the answer to this is LaTeX and Sweave, is it difficult to use a text editor such as Tinn-R or would you strongly recommend I leave behind Tinn and move over to an editor that has more LaTeX help? In reading over Friedrich Leisch's Sweave User Manual (v 1.6.0) I am beginning to think I can do everything I need with my simple editor. Before spending many hours going down that path, I thought it prudent to ask the R community. It is likely I am misunderstanding some of the process here and any clarifications are welcome. Thank you in advance for any thoughts. Tim Howard __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] [R-sig-Geo] Connecting to PostgreSQL/PostGIS from R (rgdal?)
I would also recommend you take a look at RODBC for general purpose database access. My impression is that it has received a lot more maintenance attention lately. I can't comment on the windows gdal binaries as I'm not too familiar with that platform. THK On 5/30/07, Mike Leahy [EMAIL PROTECTED] wrote: Hello, I've been trying every now and then to find a cross operating system solution that would let me access PostgreSQL (and PostGIS) from R, or to access R from PostgreSQL. I know of pl/r, which accomplishes the latter, but has yet to be successfully ported to Windows. Similarly, I've tried to use Rdbi and DBI, but I haven't had luck with those on Windows either for connecting to PostgreSQL from R. Can anyone suggest a solution for this? It would seem that rgdal could also help me in this case. Unfortunately, the version of the GDAL library that is included in the rdgal binary available on CRAN (for windows) doesn't include the PostgreSQL driver for OGR (i.e., it's not listed by the ogrDrivers() function). I compiled rgdal on Windows myself using the GDAL library from FWTools-1.3.1, but I was unsuccessful at creating a proper binary package for R. I was only able to get it to work by substituting the rgdal.dll that was installed by CRAN with the one that I compiled that links against the GDAL library from FWTools. Even though it works (at first glance with ogrInfo(), and readOGR()), I still get a warning message when I load the libary: DLL attempted to change FPU control word from 8001f to 9001f. So my question with respect to rgdal is a) is it likely that an rgdal package is going to be released in the future with the PostgreSQL driver included in GDAL/OGR, or b) are there any suggestions/instructions that might get me through the compilation and packaging process for rgdal with better success? Thanks in advance for any help, Mike ___ R-sig-Geo mailing list [EMAIL PROTECTED] https://stat.ethz.ch/mailman/listinfo/r-sig-geo -- Timothy H. Keitt, University of Texas at Austin Contact info and schedule at http://www.keittlab.org/tkeitt/ Reprints at http://www.keittlab.org/tkeitt/papers/ ODF attachment? See http://www.openoffice.org/ __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] opinions please: text editors and reporting/Sweave?
One minor warning regarding LaTeX: I have encountered journals in the psychological field (specifically, journals of the psychonomic society) that refuse to accept articles prepared in LaTeX, even if they are submitted as PDF. I'm a big LaTeX fan myself, so I really can't comprehend this. On 30-May-07, at 6:07 PM, Greg Snow wrote: Tim, First, I personnally am a big fan of LaTeX, Emacs, and ESS and I think that in the long run you would benefit from learning all of them (probably start with Emacs, then ESS, then LaTeX once you already have a knowledge of Emacs and how it can help). Since you asked about the simplest way to go, you may want to look at the odfWeave package. This gives you the power of Sweave, but without having to learn LaTeX. It works with documents in openoffice (a free office suite similar to and mostly compatible with microsoft office (word)). Using this you can create your template using openoffice writer (or MS word, then convert using openoffice), run it through R/odfWeave, and have the result as another openoffice document that can then be converted to MS word or pdf. Personally, if I am doing something for myself, or in which the output format does not matter, then I use Sweave with LaTeX (using Emacs and ESS). But, often my results need to be sent to a client that will cut and past my results into an MS word document or power point presentation. Then I find it easier to use openoffice and odfWeave and have the end result be an MS word document that can be e-mailed to the client for them to mangle in ways they feel the need to. (there is also an HTML driver for Sweave I nthe the R2HTML package where you can have the template file resemble html and the final output is in html) Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tim Howard Sent: Wednesday, May 30, 2007 2:43 PM To: r-help@stat.math.ethz.ch Subject: [R] opinions please: text editors and reporting/Sweave? dear all - I currently use Tinn-R as my text editor to work with code that I submit to R, with some output dumped to text files, some images dumped to pdf. (system: Windows 2K and XP, R 2.4.1 and R 2.5). We are using R for overnight runs to create large output data files for GIS, but then I need simple output reports for analysis results for each separate data set. Thus, I create many reports of the same style, but just based on different input data. I am recognizing that I need a better reporting system, so that I can create clean reports for each separate R run. This obviously means using Sweave and some implementation of LaTex, both of which are new to me. I've installed MikTex and successfully completed a demo or two for creating pdfs from raw LaTeX. It appears that if I want to ease my entry into the world of LaTeX, I might need to switch editors to something like Emacs (I read somewhere that Emacs helps with the TeX markup?). After quite a while wallowing at the Emacs site, I am finding that ESS is well integrated with R and might be the way to go. Aaaagh... I'm in way over my head! My questions: What, in your opinion, is the simplest way to integrate text and graphics reports into a single report such as a pdf file. If the answer to this is LaTeX and Sweave, is it difficult to use a text editor such as Tinn-R or would you strongly recommend I leave behind Tinn and move over to an editor that has more LaTeX help? In reading over Friedrich Leisch's Sweave User Manual (v 1.6.0) I am beginning to think I can do everything I need with my simple editor. Before spending many hours going down that path, I thought it prudent to ask the R community. It is likely I am misunderstanding some of the process here and any clarifications are welcome. Thank you in advance for any thoughts. Tim Howard __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code. -- Mike Lawrence Graduate Student, Department of Psychology, Dalhousie University Website: http://myweb.dal.ca/mc973993 Public calendar: http://icalx.com/public/informavore/Public The road to wisdom? Well, it's plain and simple to express: Err and err and err again, but less and less and less. - Piet Hein __ R-help@stat.math.ethz.ch
[R] determining a parent function name
Hi, All, I'm writing a wrapper for stop that produces a popup window using tcltk. Something like: error - function(...) { msg - paste(..., sep = ) if(!length(msg)) msg - if(require(tcltk, quiet = TRUE)) { tt - tktoplevel() tkwm.title(tt, Error) tkmsg - tktext(tt, bg = white) tkinsert(tkmsg, end, sprintf(Error in %s: %s, ???, msg)) tkconfigure(tkmsg, state = disabled, font = Tahoma 12, width = 50, height = 3) tkpack(tkmsg, side = bottom, fill = y) } stop(msg) } But, I would like to know from which function error() is called. For example, if I have foo - function() stop() bar - function() error() foo() Error in foo() : bar() Error in error() : and in the tk window I get Error in ???: I need the output of bar (in the tk window only) to be Error in bar(): then it's clear where error is called. I'm not worried about the output bar() produces on the console. Hope this makes sense. Thanks, --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 and provide commented, minimal, self-contained, reproducible code.
[R] runif with weights
Dear all, I would like to generate 25 numbers from 1 to 100 but I would like to have some numbers that could be more probable to come out. I was thinking of the function runif: runif(25, 1, 100) , but I don´t know how to give more weight to some numbers. Example: each number from 2 to 10 has the probability of 40% to come out but the probability of each number from 11 to 100 to come out is 60%. Kind regards and thanks in advance, João Fadista __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] runif with weights
Not sure why you have set the probability of a 1 to 0 but maybe something like this might be what you want: round( ifelse( rbinom(25, 1, 0.4), runif(25, 2, 10), runif(25, 11, 100) ) ) [1] 2 6 34 90 79 71 83 8 47 36 21 32 17 71 3 16 9 65 94 6 30 5 7 10 13 I would like to generate 25 numbers from 1 to 100 but I would like to have some numbers that could be more probable to come out. I was thinking of the function runif: runif(25, 1, 100) , but I don´t know how to give more weight to some numbers. Example: each number from 2 to 10 has the probability of 40% to come out but the probability of each number from 11 to 100 to come out is 60%. -- Ken Knoblauch Inserm U846 Institut Cellule Souche et Cerveau Département Neurosciences Intégratives 18 avenue du Doyen Lépine 69500 Bron France tel: +33 (0)4 72 91 34 77 fax: +33 (0)4 72 91 34 61 portable: +33 (0)6 84 10 64 10 http://www.pizzerialesgemeaux.com/u846/ __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] runif with weights
You did not explicitly say it, but your example indicates that you want to sample from integers only (else what would weights mean?). So... ?sample -- in particular note the prob argument and read help docs carefully e.g. sample(100,25,prob=c(0,rep.int(.4,9),rep.int(.6,90))) ## without replacement Bert Gunter Genentech Nonclinical Statistics -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ken Knoblauch Sent: Wednesday, May 30, 2007 5:59 PM To: [EMAIL PROTECTED] Cc: r-help@stat.math.ethz.ch Subject: [R] runif with weights Not sure why you have set the probability of a 1 to 0 but maybe something like this might be what you want: round( ifelse( rbinom(25, 1, 0.4), runif(25, 2, 10), runif(25, 11, 100) ) ) [1] 2 6 34 90 79 71 83 8 47 36 21 32 17 71 3 16 9 65 94 6 30 5 7 10 13 I would like to generate 25 numbers from 1 to 100 but I would like to have some numbers that could be more probable to come out. I was thinking of the function runif: runif(25, 1, 100) , but I don´t know how to give more weight to some numbers. Example: each number from 2 to 10 has the probability of 40% to come out but the probability of each number from 11 to 100 to come out is 60%. -- Ken Knoblauch Inserm U846 Institut Cellule Souche et Cerveau Département Neurosciences Intégratives 18 avenue du Doyen Lépine 69500 Bron France tel: +33 (0)4 72 91 34 77 fax: +33 (0)4 72 91 34 61 portable: +33 (0)6 84 10 64 10 http://www.pizzerialesgemeaux.com/u846/ __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] sizing and saving graphics in R
Dear R wizards, I am seeking advice on graphics in R. Specifically, how to manipulate the size and save a plot I have produced using the LDheatmap library. I confess I am relatively new to graphics in R, but I would greatly appreciate any suggestions you may have. LDheatmap produces a coloured triangular matrix of pairwise associations between 600 genetic markers in my dataset. Initially the graphical output was confined to the computer screen, such that each pairwise marker association was displayed as approximately 1 pixel (too small for me to interpret). I have successfully managed to play with the LDheatmap function to enlarge the size of viewport by changing the following code in LDheatmap #From heatmapVP - viewport(width = unit(0.8, snpc), height = unit(0.8, snpc), name=vp.name) #To heatmapVP - viewport(width = unit(25, inches), height = unit(25, inches), name=vp.name) This produces a much larger plot (so big that the majority is not seen on the screen). I would like to save the entire thing so that I can import it into photoshop or some other image software. My problem is that when I save using the R graphics console (File-Save As-bmp), it only saves the section I can see on the screen. Any suggestions on how to save the whole plot or manipulate the plot so I get higher resolution would be much appreciated. Thanks for your help in advance, Felicity. ___ Dr Felicity Jones Department of Developmental Biology Stanford University School of Medicine Beckman Center 279 Campus Drive Stanford CA 94305-5329 USA __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] sizing and saving graphics in R
I use the savePlot function for saving graphics. The following will save the active graphics panel in your working directory, in format wmf, which I find has a high resolution. Check out other possible formats in help. savePlot(filename = myfilename,type = c(wmf)) Murray On 31/05/07, Felicity Jones [EMAIL PROTECTED] wrote: Dear R wizards, I am seeking advice on graphics in R. Specifically, how to manipulate the size and save a plot I have produced using the LDheatmap library. I confess I am relatively new to graphics in R, but I would greatly appreciate any suggestions you may have. LDheatmap produces a coloured triangular matrix of pairwise associations between 600 genetic markers in my dataset. Initially the graphical output was confined to the computer screen, such that each pairwise marker association was displayed as approximately 1 pixel (too small for me to interpret). I have successfully managed to play with the LDheatmap function to enlarge the size of viewport by changing the following code in LDheatmap #From heatmapVP - viewport(width = unit(0.8, snpc), height = unit(0.8, snpc), name=vp.name) #To heatmapVP - viewport(width = unit(25, inches), height = unit(25, inches), name=vp.name) This produces a much larger plot (so big that the majority is not seen on the screen). I would like to save the entire thing so that I can import it into photoshop or some other image software. My problem is that when I save using the R graphics console (File-Save As-bmp), it only saves the section I can see on the screen. Any suggestions on how to save the whole plot or manipulate the plot so I get higher resolution would be much appreciated. Thanks for your help in advance, Felicity. ___ Dr Felicity Jones Department of Developmental Biology Stanford University School of Medicine Beckman Center 279 Campus Drive Stanford CA 94305-5329 USA __ 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 and provide commented, minimal, self-contained, reproducible code. -- Murray Pung Statistician, Datapharm Australia Pty Ltd 0404 273 283 [[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 and provide commented, minimal, self-contained, reproducible code.
[R] cox goodness of fit
Is there an implementation of the Cox-Snell residuals / Nelson-Aalen plot for goodness of fit? Or otherwise is there an appropriate Goodness of Fit diagnostic? Thanks Murray -- Murray Pung Statistician, Datapharm Australia Pty Ltd 0404 273 283 [[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 and provide commented, minimal, self-contained, reproducible code.
[R] how to preserve trained model in LDA?
Hi all: I'm developing an application in which I use standard data to train the model in LDA and use the trained model to predict on test data. I can't train the model every time when I do prediction. So I need to save the trained model onto disk after the first training. Does anybody have idea about this? You help is highly appreciated. Best Regards Feng [[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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] opinions please: text editors and reporting/Sweave?
Tim Howard wrote: dear all - I currently use Tinn-R as my text editor to work with code that I submit to R, with some output dumped to text files, some images dumped to pdf. (system: Windows 2K and XP, R 2.4.1 and R 2.5). We are using R for overnight runs to create large output data files for GIS, but then I need simple output reports for analysis results for each separate data set. Thus, I create many reports of the same style, but just based on different input data. I am recognizing that I need a better reporting system, so that I can create clean reports for each separate R run. This obviously means using Sweave and some implementation of LaTex, both of which are new to me. I've installed MikTex and successfully completed a demo or two for creating pdfs from raw LaTeX. It appears that if I want to ease my entry into the world of LaTeX, I might need to switch editors to something like Emacs (I read somewhere that Emacs helps with the TeX markup?). After quite a while wallowing at the Emacs site, I am finding that ESS is well integrated with R and might be the way to go. Aaaagh... I'm in way over my head! If you are used to Windows, you might find the shareware editors WinEdt or Textpad more familiar. WinEdt has advantages of lots of LaTeX integration. Duncan Murdoch My questions: What, in your opinion, is the simplest way to integrate text and graphics reports into a single report such as a pdf file. If the answer to this is LaTeX and Sweave, is it difficult to use a text editor such as Tinn-R or would you strongly recommend I leave behind Tinn and move over to an editor that has more LaTeX help? In reading over Friedrich Leisch's Sweave User Manual (v 1.6.0) I am beginning to think I can do everything I need with my simple editor. Before spending many hours going down that path, I thought it prudent to ask the R community. It is likely I am misunderstanding some of the process here and any clarifications are welcome. Thank you in advance for any thoughts. Tim Howard __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] opinions please: text editors and reporting/Sweave?
Winshell (http://www.winshell.de/) is another (free) option if you want a Windows editor with good MikTEX integration. On 5/31/07, Duncan Murdoch [EMAIL PROTECTED] wrote: Tim Howard wrote: dear all - I currently use Tinn-R as my text editor to work with code that I submit to R, with some output dumped to text files, some images dumped to pdf. (system: Windows 2K and XP, R 2.4.1 and R 2.5). We are using R for overnight runs to create large output data files for GIS, but then I need simple output reports for analysis results for each separate data set. Thus, I create many reports of the same style, but just based on different input data. I am recognizing that I need a better reporting system, so that I can create clean reports for each separate R run. This obviously means using Sweave and some implementation of LaTex, both of which are new to me. I've installed MikTex and successfully completed a demo or two for creating pdfs from raw LaTeX. It appears that if I want to ease my entry into the world of LaTeX, I might need to switch editors to something like Emacs (I read somewhere that Emacs helps with the TeX markup?). After quite a while wallowing at the Emacs site, I am finding that ESS is well integrated with R and might be the way to go. Aaaagh... I'm in way over my head! If you are used to Windows, you might find the shareware editors WinEdt or Textpad more familiar. WinEdt has advantages of lots of LaTeX integration. Duncan Murdoch My questions: What, in your opinion, is the simplest way to integrate text and graphics reports into a single report such as a pdf file. If the answer to this is LaTeX and Sweave, is it difficult to use a text editor such as Tinn-R or would you strongly recommend I leave behind Tinn and move over to an editor that has more LaTeX help? In reading over Friedrich Leisch's Sweave User Manual (v 1.6.0) I am beginning to think I can do everything I need with my simple editor. Before spending many hours going down that path, I thought it prudent to ask the R community. It is likely I am misunderstanding some of the process here and any clarifications are welcome. Thank you in advance for any thoughts. Tim Howard __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code. [[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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Generating Data using Formulas
Christian, The formula language is not suited to such recursive useage AFAICS. You can _vectorize_ your code like this: cmat - outer( 1:25, 1:25, function(y,x) ifelse( xy, 0, 0.8^(y-x) ) ) res - replicate(1000,{ y - 1 + cmat %*% rnorm(25) coef(lm(y[-1]~y[-25])) }) rowMeans(res) # mean of 1000 replicates HTH, Chuck On Tue, 29 May 2007, Chrisitan Falde wrote: Hello, My name is Christian Falde. I am new to R. My problem is this. I am attempting to learn R on my own. In so doing I am using some problems from Davidson and MacKinnon Econometric Theory and Methods to do so. This is because I can already do the some of the problems in SAS so I am attempting to rework them using R. Seemed logical to me, now I am stuck and its really bugging me. The problem is this Generate a data set sample size of 25 with the formula y=1+.8*y(t-1)+ u. Where y is the dependent, y(t-1) is the dependent variable lagged one peroid, and u is the classical error term. Assume y0=0 and the u is NID(0,1). Use this sample to compute the OLS estimates B1 (1) and B2(.8). Repeat at least 100 times and find the average of the B's. Use these average to estimate the bias of the ols estimators. To start I did the following non lagged program. final-function(i,j){x-function(i) {10*i} y-function(i,j) {1+.8*10*i+100*rnorm(j)} datathreeone- data.frame(replicate(100,coef(lm(y(i,j)~x(i) rowMeans(datathreeone)} final(1:25,25) final(1:50,50) final(1:100,100) final(1:200,200) final(1:1,1) Now the only thing I need to to is change .8*10*i which is exogenous to .8* y(t-1) . There are two reasons why I did it this way. I needed the rnorm(i) to generate a new set of u's each replication, and I wanted to be able to use the function as i did to make the results more concise. For the lag in SAS we used an if then else logic relating to the number of observation. This in R would have to be linked to the invisable row number. I think I need an index variable for the row. Perhaps, sorry thinking while typing. Another reason why I am stuck, the lag function was seemingly straight forward. lag (x, k=1) yet x has to be a matrix so when I tried to do it like above with y as a function R complained. I have been working on this for a couple of days now so everything is begining to not make sense. It just seems to me to get the matrix to work out I would need to have two matrices. dependentand explanatory y1 = sum ( 1 +.8*0 + 100*rnorm(i)) y2 = sum ( 1 +.8* (dependent row 1) + 100*rnorm(i)) etc I just am not sure how to do that. Please help and thank you for your time, christian falde [snip] Charles C. Berry(858) 534-2098 Dept of Family/Preventive Medicine E mailto:[EMAIL PROTECTED] UC San Diego http://biostat.ucsd.edu/~cberry/ La Jolla, San Diego 92093-0901 __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Factor function: odd behavior when labels argument containsduplicates?
There is a difference between levels and labels. I think this is what you want. x - factor(rep(0:5, 2)) x [1] 0 1 2 3 4 5 0 1 2 3 4 5 Levels: 0 1 2 3 4 5 levels(x) - c(1,1:5) x [1] 1 1 2 3 4 5 1 1 2 3 4 5 Levels: 1 2 3 4 5 table(x) x 1 2 3 4 5 4 2 2 2 2 Bill Venables CSIRO Laboratories PO Box 120, Cleveland, 4163 AUSTRALIA Office Phone (email preferred): +61 7 3826 7251 Fax (if absolutely necessary): +61 7 3826 7304 Mobile:(I don't have one!) Home Phone: +61 7 3286 7700 mailto:[EMAIL PROTECTED] http://www.cmis.csiro.au/bill.venables/ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steen Ladelund Sent: Wednesday, 30 May 2007 6:27 PM To: r-help@stat.math.ethz.ch Subject: [R] Factor function: odd behavior when labels argument containsduplicates? Hi all. I think it would be nice to be able to combine levels of a factor on creation a la x - rep(0:5,5) y - factor(x,levels=0:5,labels=c('1','1',2:5)) ## (1) y [1] 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 Levels: 1 1 2 3 4 5 I thougt this would (should?) create a factor with 5 levels, ie combining 0 and 1 in x into one level in y. I find it hard to predict the behavior of the following lines: table(factor(y)) ## Odd ? 1 1 2 3 4 5 10 0 5 5 5 5 table(factor(factor(y))) ## This is what I want 1 2 3 4 5 10 5 5 5 5 It also seems strange that this should be the way to go: levels(y) - levels(y) ## Does what I want following (1) even tough it appear to be a void statement? y y [1] 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 1 1 2 3 4 5 Levels: 1 2 3 4 5 Am I missing an important point about factors or the factor function? steen Steen Ladelund, statistician +4543233275 [EMAIL PROTECTED] Research Center for Prevention and Health Glostrup University Hospital, Denmark www.fcfs.kbhamt.dk __ 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 and provide commented, minimal, self-contained, reproducible code. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] lattice: aligning independent graphs
On Sun, 27 May 2007 15:09:43 -0700, Deepayan Sarkar [EMAIL PROTECTED] wrote: [...] Better in what sense? You are trying to use lattice to do something that lattice isn't designed to do, so there is no solution that is clean in a philosophical sense. Ideally, you want something like grid, that allows you to control the layout in as much detail as you want (you can still use panel.xyplot, panel.grid, etc. as panel functions once you set up everything else). On the other hand, lattice gets you very close, and the only improvement I can think of is some sort of automated scheme to align the x labels to the respective columns. Here's one approach to do that (taking advantage of the fact that xlab can be an arbitrary grob): [...] I'm not yet savvy with grid, but if one were to do the same for y axis labels, I thought the following modification to your code might do that: --cut here---start- myXlabGrob - function(...) ## ...is lab1, lab2, etc { ## you can add arguments to textGrob for more control ## in the next line labs - lapply(list(...), textGrob, rot=90) nlabs - length(labs) lab.heights - lapply(labs, function(lab) unit(1, grobheight, data=list(lab))) lab.layout - grid.layout(ncol=1, nrow=nlabs, heights=unit(1, null), widths=do.call(max, lab.heights), respect=TRUE) lab.gf - frameGrob(layout=lab.layout) for (i in seq_len(nlabs)) { lab.gf - placeGrob(lab.gf, labs[[i]], row=i, col=1) } lab.gf } xyplot(1:9 ~ 1:9 | gl(3, 1, 9), layout=c(1, 3), ylab=myXlabGrob('Trial number', 'Subject number', 'Experimental condition'), strip=FALSE) --cut here---end--- which gives wrong width and placement. How can this be modified so it places the labels close to the axis annotation, centered on each panel? Thanks in advance. Cheers, -- Seb __ 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 and provide commented, minimal, self-contained, reproducible code.