Re: [R] is it possible to form matrix of matrices...and multiple arrays
booop booop a écrit : 1...Kindly tell me is it possible to form a matrix which contains a no of matrices.. for eg.. if a,b,c,d are matrices and e is a matrix which contains a,b,c,d as rows and columns.. I don't think you can use matrix() to store other matrix() inside. But array() is a solution to store matrix() inside. (At least I have use it). __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Plot Data Points in boxplots
Paul Murrell a écrit : You might also like to take a look at Figure 3.27 (and associated R code) on http://www.stat.auckland.ac.nz/~paul/RGraphics/chapter3.html Thank you very much for making this page available. The power of R graphics, and the number of options, makes such examples very useful. (At least for beginners like me !) 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
Re: [R] Plot Data Points in boxplots
vincent == vincent [EMAIL PROTECTED] on Thu, 29 Sep 2005 08:30:10 +0200 writes: vincent Paul Murrell a écrit : You might also like to take a look at Figure 3.27 (and associated R code) on http://www.stat.auckland.ac.nz/~paul/RGraphics/chapter3.html vincent Thank you very much for making this page available. vincent The power of R graphics, and the number of options, vincent makes such examples very useful. (At least for vincent beginners like me !) Thanks. To really thank Paul, you should buy the book to which the above web page refers : http://www.stat.auckland.ac.nz/~paul/RGraphics/rgraphics.html A very nice and useful book to have, indeed! Martin Maechler, ETH Zurich __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] p-level in packages mgcv and gam
Yves Magliulo said the following on 2005-09-28 17:05: hi, i'll try to help you, i send a mail about this subject last week... and i did not have any response... I'm using gam from package mgcv. 1) How to interpret the significance of smooth terms is hard for me to understand perfectly : using UBRE, you fix df. p-value are estimated by chi-sq distribution using GCV, the best df are estimated by GAM. (that's what i want) and p-values This is not correct. The df are estimated in both cases (i.e. UBRE and GCV), but the scale parameter is fixed in the UBRE case. Hence, by default UBRE is used for family = binomial or poisson since the scale parameter is assumed to be 1. Similarly, GCV is the default for family = gaussian since we most often want the scale (usually denoted sigma^2) to be estimated. are estimated by an F distribution But in that case they said use at your own risk in ?summary.gam The warning applies in both cases. The p-values are conditional on the smoothing parameters, and the uncertainty of the smooths is not taken into account when computing the p-values. so you can also look at the chi.sq : but i don't know how to choose a No... criterion like for p-values... for me, chi.sq show the best predictor in a model, but it's hard to reject one with it. Which version of mgcv do you use? The confusion probably stems from earlier versions of mgcv ( 1.3-5): the summary and anova methods used to have a column denoted Chi.sq even when the displayed statistic was computed as F. Recent versions of mgcv has summary(b) Family: gaussian Link function: identity Formula: y ~ s(x0) + s(x1) + s(x2) + s(x3) Parametric coefficients: Estimate Std. Error t value Pr(|t|) (Intercept) 7.9150 0.1049 75.44 2e-16 *** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Approximate significance of smooth terms: edf Est.rank F p-value s(x0) 5.1739.000 3.785 0.000137 *** s(x1) 2.3579.000 34.631 2e-16 *** s(x2) 8.5179.000 84.694 2e-16 *** s(x3) 1.0001.000 0.444 0.505797 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 R-sq.(adj) = 0.726 Deviance explained = 73.7% GCV score = 4.611 Scale est. = 4.4029n = 400 If we assume that the scale is known and fixed at 4.4029, we get summary(b, dispersion = 4.4029) Family: gaussian Link function: identity Formula: y ~ s(x0) + s(x1) + s(x2) + s(x3) Parametric coefficients: Estimate Std. Error z value Pr(|z|) (Intercept) 7.9150 0.1049 75.44 2e-16 *** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Approximate significance of smooth terms: edf Est.rank Chi.sq p-value s(x0) 5.1739.000 34.067 8.7e-05 *** s(x1) 2.3579.000 311.679 2e-16 *** s(x2) 8.5179.000 762.255 2e-16 *** s(x3) 1.0001.000 0.444 0.505 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 R-sq.(adj) = 0.726 Deviance explained = 73.7% GCV score = 4.611 Scale est. = 4.4029n = 400 Note that t/F changed into z/Chi.sq. so as far as i m concerned, i use GCV methods, and fix a 5% on the null hypothesis (pvalue) to select significant predictor. after, i look at my smooth, and if the parametrization look fine to me, i validate. generaly, for p-values smaller than 0.001, you can be confident. over 0.001, you have to check. 2) for difference between package gam and mgcv, i sent a mail about this The underlying algorithms are very different. HTH, Henric one year ago, here's the response : - package gam is based very closely on the GAM approach presented in Hastie and Tibshirani's Generalized Additive Models book. Estimation is by back-fitting and model selection is based on step-wise regression methods based on approximate distributional results. A particular strength of this approach is that local regression smoothers (`lo()' terms) can be included in GAM models. - gam in package mgcv represents GAMs using penalized regression splines. Estimation is by direct penalized likelihood maximization with integrated smoothness estimation via GCV or related criteria (there is also an alternative `gamm' function based on a mixed model approach). Strengths of the this approach are that s() terms can be functions of more than one variable and that tensor product smooths are available via te() terms - these are useful when different degrees of smoothness are appropriate relative to different arguments of a smooth. (...) Basically, if you want integrated smoothness selection, an underlying parametric representation, or want smooth interactions in your models then mgcv is probably worth a try (but I would say that). If you want to use local regression smoothers and/or prefer the stepwise selection approach then package gam is for you. i think the difference of p-values between :gam and :mgcv, is because you don't have
[R] how to fix the level-1 variances in lme()?
Dear all, Edmond Ng (http://multilevel.ioe.ac.uk/softrev/reviewsplus.pdf) provides an example to fit the mixed effects meta-analysis in Splus 6.2. The syntax is: lme(fixed=d~wks, data=meta, random=~1|study, weights=varFixed(~Vofd), control=lmeControl(sigma=1)) where d is the effect size, study is the study number, Vofd is the variance of the effect size and meta is the data frame. sigma=1 is required to constrain the level 1 variance in applying mixed-effects models in meta-analysis. In Splus 6.1, I found that the help manual of nlme includes sigma as an optional argument in lmeControl() to fix the within-group standard error during the optimization. However, both Pinheiro and Bates (2000, p.476) and the help manual of the nlme package in R for Version 3.1-65 (http://cran.r-project.org/doc/packages/nlme.pdf, p.172) do not include sigma as an argument for lmeControl(). I would like to know how I could fix the level-1 variances as known values in lme(). Thanks in advance! Best, Mike -- - Mike W.L. Cheung Phone: (852) 2857-8621 Department of Psychology Fax: (852) 2858-3518 The University of Hong KongE-mail: [EMAIL PROTECTED] HONG KONG Website: http://web.hku.hk/~mikewlch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] how to fix the level-1 variances in lme()?
On Thu, 29 Sep 2005, Mike Cheung wrote: Dear all, Edmond Ng (http://multilevel.ioe.ac.uk/softrev/reviewsplus.pdf) provides an example to fit the mixed effects meta-analysis in Splus 6.2. The syntax is: lme(fixed=d~wks, data=meta, random=~1|study, weights=varFixed(~Vofd), control=lmeControl(sigma=1)) where d is the effect size, study is the study number, Vofd is the variance of the effect size and meta is the data frame. sigma=1 is required to constrain the level 1 variance in applying mixed-effects models in meta-analysis. In Splus 6.1, I found that the help manual of nlme includes sigma as an optional argument in lmeControl() to fix the within-group standard error during the optimization. However, both Pinheiro and Bates (2000, p.476) and the help manual of the nlme package in R for Version 3.1-65 (http://cran.r-project.org/doc/packages/nlme.pdf, p.172) do not include sigma as an argument for lmeControl(). I would like to know how I could fix the level-1 variances as known values in lme(). Use S-PLUS and nlme = 3.3 (as I recall). R's and S-PLUS's nlme are not the same (they diverged from a common base around 1998). It is a while (2 years?) since I needed this capability, but at the time it was not possible in R's lme. Since you have access to S-PLUS, why not use it? -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] is it possible to form matrix of matrices...and multiple arrays
On Thu, 29 Sep 2005 [EMAIL PROTECTED] wrote: booop booop a écrit : 1...Kindly tell me is it possible to form a matrix which contains a no of matrices.. for eg.. if a,b,c,d are matrices and e is a matrix which contains a,b,c,d as rows and columns.. I don't think you can use matrix() to store other matrix() inside. But array() is a solution to store matrix() inside. (At least I have use it). You _can_ do this with matrix() (although that was not quite what was asked). Try a - b - c - d - matrix(1:4, 2, 2) e - matrix(list(a,b,c,d), 2,2) e e[1,2][[1]] -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] priceIts
On Thu, 29 Sep 2005, Mulholland, Tom wrote: Well I downloaded the data using the link in your message which suggests that the code is right. I don't have its loaded (I assume it's from the irregular time series package) The package is called its, as the message correctly said. so I can't test the code as you have it. chart.yahoo.com is notoriously fickle (as we have seen from the tseries example using it). Note that the first command (which gave me no error) also downloaded from there, so this is not going to be an access issue. I do get a 404 error on that URL (from two separate ISPs), but not if gdax is twice replaced by ftse. I do think the posting guide makes clear this is an issue for the package maintainer, not for R-help. Have you tried checking to see it it is an issue with the internet (your browser, blocked sites etc) rather than a problem in R? Tom -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Remigijus Lapinskas Sent: Thursday, 29 September 2005 12:31 PM To: r-help@stat.math.ethz.ch Subject: [R] priceIts Dear All, There is an example for the priceIts function (the its package) which does not work for me as expected. ?priceIts x1 - priceIts(instrument = c(^ftse), start = 1998-01-01, + quote = Close) Error in validObject(.Object) : invalid class its object: Missing values in dates x2 - priceIts(instrument = c(^gdax), start = 1998-01-01, + quote = Close) Error in download.file(url, destfile, method = method, quiet = quiet) : cannot open URL 'http://chart.yahoo.com/table.csv?s=^gdaxa=0b=01c=1998d=8; e=28f=2005g=dq=qy=0z=^gdaxx=.csv' In addition: Warning message: cannot open: HTTP status was '404 Not Found' x - union(x1,x2) Error: Object x1 not found Error in union(x1, x2) : unable to find the argument 'x' in selecting a method for function 'union' names(x) - c(FTSE,DAX) plot(x,lab=TRUE) Any help appreciated. Best wishes, Remigijus __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] multiple plots on same x axis
Hi You can use e.g. plot - points construction plot(Day, gene1) points(Day, gene2, col=2) see ?points ?lines You also need to set appropriate y range by ylim argument to first plot. And if you used search facility in CRAN and asked it multiple plots on same x axis you would have got many other posibilities for your question. HTH Petr On 27 Sep 2005 at 15:34, IAIN GALLAGHER wrote: Date sent: Tue, 27 Sep 2005 15:34:53 +0100 (BST) From: IAIN GALLAGHER [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Subject:[R] multiple plots on same x axis Hi. I have two vectors of gene expression for each of several days. I want to plot both vectors on the same plot for a visual representation of up versus down regulation. I've tried using add=T but that doesn't work. eg plot(Day, gene1) plot(Day, gene2, add=T) Any help would be appreciated. Iain __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] is it possible to form matrix of matrices...and multiple arrays
Prof Brian Ripley a écrit : On Thu, 29 Sep 2005 [EMAIL PROTECTED] wrote: I don't think you can use matrix() to store other matrix() inside. But array() is a solution to store matrix() inside. (At least I have use it). You _can_ do this with matrix() (although that was not quite what was asked). Try a - b - c - d - matrix(1:4, 2, 2) e - matrix(list(a,b,c,d), 2,2) sorry for the wrong answer, and thanks for the correction. (I certainly do not use list() as much as it should be.) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] how to fix the level-1 variances in lme()?
Mike I do not believe this is availabe in either lme or lmer in R, only S-Plus. -Original Message- From: [EMAIL PROTECTED] on behalf of Mike Cheung Sent: Thu 9/29/2005 4:32 AM To: r-help@stat.math.ethz.ch Cc: Subject:[R] how to fix the level-1 variances in lme()? Dear all, Edmond Ng (http://multilevel.ioe.ac.uk/softrev/reviewsplus.pdf) provides an example to fit the mixed effects meta-analysis in Splus 6.2. The syntax is: lme(fixed=d~wks, data=meta, random=~1|study, weights=varFixed(~Vofd), control=lmeControl(sigma=1)) where d is the effect size, study is the study number, Vofd is the variance of the effect size and meta is the data frame. sigma=1 is required to constrain the level 1 variance in applying mixed-effects models in meta-analysis. In Splus 6.1, I found that the help manual of nlme includes sigma as an optional argument in lmeControl() to fix the within-group standard error during the optimization. However, both Pinheiro and Bates (2000, p.476) and the help manual of the nlme package in R for Version 3.1-65 (http://cran.r-project.org/doc/packages/nlme.pdf, p.172) do not include sigma as an argument for lmeControl(). I would like to know how I could fix the level-1 variances as known values in lme(). Thanks in advance! Best, Mike -- - Mike W.L. Cheung Phone: (852) 2857-8621 Department of Psychology Fax: (852) 2858-3518 The University of Hong KongE-mail: [EMAIL PROTECTED] HONG KONG Website: http://web.hku.hk/~mikewlch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Plot Data Points in boxplots
Martin Maechler a écrit : To really thank Paul, you should buy the book to which the above web page refers : http://www.stat.auckland.ac.nz/~paul/RGraphics/rgraphics.html A very nice and useful book to have, indeed! I just broked my piggy bank for Peter Dalgaard's one, so I'll have to wait a bit before it blows out again. (Peter Dalgaard Introductory Statistics With R http://www.amazon.fr/exec/obidos/ASIN/0387954759/qid=1127989069/sr=1-1/ref=sr_1_0_1/171-5704479-0348209 , also very useful as an introduction.) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] p-level in packages mgcv and gam
hi, thanks for reply. 1) sorry for the mistake, (GCV and UBRE estimate df in both case in deed) i've made a confusion. 2) i use mgcv 1.1-8. (!) so in deed again, i need an update. my questions : 3) i can see in the last version of summary.gam an estimated rank: how do i interpret this? 4) I'm still confusing with this question : should i look at R-sq rather than Deviance explain, or both? 5) How can i estimate numericaly the contribution of each smooth against the others. In others words, is there a way to quantify significance like a percentage of how the model is improved by each of my smooth? 6) mgcv : gam and gam : gam can't be compared. that's 2 distinct approach of gam, isn't it? for Denis : 7) to validate the significance of a smooth where p-values is betwwen 0.01 and 0.05, there is no perfect rules... you have to decide according to your need. look at df, the se of your smooth, how your model is improved, if the parametrization look fine according of what you want/know of your predictor etc... but that's my point of view, maybe someone have Criterion more precise. Yves Le jeu 29/09/2005 à 09:55, Henric Nilsson a écrit : Yves Magliulo said the following on 2005-09-28 17:05: hi, i'll try to help you, i send a mail about this subject last week... and i did not have any response... I'm using gam from package mgcv. 1) How to interpret the significance of smooth terms is hard for me to understand perfectly : using UBRE, you fix df. p-value are estimated by chi-sq distribution using GCV, the best df are estimated by GAM. (that's what i want) and p-values This is not correct. The df are (i.e. UBRE and GCV), but the scale parameter is fixed in the UBRE case. Hence, by default UBRE is used fsignificance of smooth termsor family = binomial or poisson since the scale parameter is assumed to be 1. Similarly, GCV is the default for family = gaussian since we most often want the scale (usually denoted sigma^2) to be estimated. are estimated by an F distribution But in that case they said use at your own risk in ?summary.gam The warning applies in both cases. The p-values are conditional on the smoothing parameters, and the uncertainty of the smooths is not taken into account when computing the p-values. so you can also look at the chi.sq : but i don't know how to choose a No... criterion like for p-values... for me, chi.sq show the best predictor in a model, but it's hard to reject one with it. Which version of mgcv do you use? The confusion probably stems from earlier versions of mgcv ( 1.3-5): the summary and anova methods used to have a column denoted Chi.sq even when the displayed statistic was computed as F. Recent versions of mgcv has summary(b) Family: gaussian Link function: identity Formula: y ~ s(x0) + s(x1) + s(x2) + s(x3) Parametric coefficients: Estimate Std. Error t value Pr(|t|) (Intercept) 7.9150 0.1049 75.44 2e-16 *** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Approximate significance of smooth terms: edf Est.rank F p-value s(x0) 5.1739.000 3.785 0.000137 *** s(x1) 2.3579.000 34.631 2e-16 *** s(x2) 8.5179.000 84.694 2e-16 *** s(x3) 1.0001.000 0.444 0.505797 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 R-sq.(adj) = 0.726 Deviance explained = 73.7% GCV score = 4.611 Scale est. = 4.4029n = 400 If we assume that the scale is known and fixed at 4.4029, we get summary(b, dispersion = 4.4029) Family: gaussian Link function: identity Formula: y ~ s(x0) + s(x1) + s(x2) + s(x3) Parametric coefficients: Estimate Std. Error z value Pr(|z|) (Intercept) 7.9150 0.1049 75.44 2e-16 *** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Approximate significance of smooth terms: edf Est.rank Chi.sq p-value s(x0) 5.1739.000 34.067 8.7e-05 *** s(x1) 2.3579.000 311.679 2e-16 *** s(x2) 8.5179.000 762.255 2e-16 *** s(x3) 1.0001.000 0.444 0.505 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 R-sq.(adj) = 0.726 Deviance explained = 73.7% GCV score = 4.611 Scale est. = 4.4029n = 400 Note that t/F changed into z/Chi.sq. so as far as i m concerned, i use GCV methods, and fix a 5% on the null hypothesis (pvalue) to select significant predictor. after, i look at my smooth, and if the parametrization look fine to me, i validate. generaly, for p-values smaller than 0.001, you can be confident. over 0.001, you have to check. 2) for difference between package gam and mgcv, i sent a mail about this The underlying algorithms are very different. HTH, Henric one year ago, here's the response : - package gam is based very closely on the GAM approach presented in Hastie and Tibshirani's Generalized
Re: [R] boxplot and xlim confusion?
Hi Karin I did not have seen any answer for your question yet so here is a try. I gues you want the horizotal layout or your boxplot. boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=letters[1:3]) boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=c(NA,b,NA)) So this is closest what I could dig from your text. ***reproducible*** example would by good starting point what you want and was not able to produce. HTH Petr On 28 Sep 2005 at 15:50, Karin Lagesen wrote: To: [EMAIL PROTECTED] From: Karin Lagesen [EMAIL PROTECTED] Date sent: Wed, 28 Sep 2005 15:50:06 +0200 Subject:[R] boxplot and xlim confusion? I have some code as shown below. Basically, I would like three boxplots to be set next to each other with no ylabels on the two inner plots, and I want the same x axis range on all three. However, it seems like boxplot does not respect the xlim setting. I've tried the various ways I thought would work (par, boxplot(...xlim=)) but none of them seem to work. I then tried plot.window, that did not work. I also have another curious question for you. With the code below I tried to call plot.new and then plot.window before each new plot. What happens then is that the first figure goes on one page whereas the two others get put on the next page with a nice big gap in the middle. Does any of you have an explanation for that? names - c( LSU, stop, LSU, start, SSU, stop, SSU, start, TSU, stop, TSU, start) elsustop - read.table(28s.euk.accuracy.stop.dev) [skipped lots of read.table, which just reads files with one number on each line] par(mfcol=c(1,3)) par(mai = c(0,0,0.5,0.2), omi = c(1,1,1,1)) xaxis = c(-6000,1000) yaxis = c(0,7) #plot.new() #plot.window(xlim=xaxis, ylim=yaxis) boxplot(alsustop$V1 ,alsustart$V1 ,assustop$V1 ,alsustart$V1 ,atsustop$V1 ,atsustart$V1 ,names=names,col=c(lightblue,orange,lightblue,orange,lightblu e,orange) ,horizontal = TRUE, main=ARC, xaxs = i, las=1) #plot.new() #plot.window(xlim=xaxis, ylim=yaxis) boxplot(blsustop$V1 ,blsustart$V1 ,bssustop$V1 ,blsustart$V1 ,btsustop$V1 ,btsustart$V1 ,col=c(lightblue,orange,lightblue,orange,lightblue,orange) ,horizontal = TRUE, main=BAC, xaxs = i) #plot.new() #plot.window(xlim=xaxis, ylim=yaxis) boxplot(elsustop$V1 ,elsustart$V1 ,essustop$V1 ,elsustart$V1 ,etsustop$V1 ,etsustart$V1 ,col=c(lightblue,orange,lightblue,orange,lightblue,orange) ,horizontal = TRUE, main=EUK, xaxs = i) Karin (sorry if you're getting sick of me..:)) -- Karin Lagesen, PhD student [EMAIL PROTECTED] http://www.cmbn.no/rognes/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Easy cut paste from Excel to R?
Sorry to revive and old topic, but writing to the clipboard seems to have a problem for me: column names are ignored. Example: # ~~~ # write.clipboard # ~~~ write.clipboard = function(obj) { write.table(obj, file(clipboard), sep=\t, row.names=F, col.names=T) } a= matrix(1:4,2,2) colnames(a) = c(a, b) write.clipboard(a) a = as.data.frame(a) write.clipboard(a) both attempts will paste the date without column names. Any idea why? Thanks, -Jose On 2/16/05, Werner Wernersen [EMAIL PROTECTED] wrote: Thank you all very much for the answers! The read.table / read.delim2 commands are exactly what I was looking for to get a couple of numbers or a little matrix quickly into R without creating an extra text file every time. And it works the other way around as well: write.table(x, file(clipboard), sep=\t) Fantastic! Thanks again, Werner Nick Drew wrote: I've had good luck with the scan() function when I want to get a few numbers from Excel into R quickly to use it as a calculator. CAVEAT: you have to have the numbers you want to copy in a column not a row in Excel. For example: In Excel your data are in a column as follows: Col A 1 2 3 Then copy the 3 cells (e.g. 1, 2,3) in Excel and open R and type in: data - scan() Then Paste using Ctrl-V. Hit the Enter key. You know have an object called data that you can use and manipulate in R. I've taken this even further by creating an R function that will take a column of numbers from Excel and then scan() them into R, create a matrix, and then perform a Chi-square test. Let me know if you'd like to know more. I'm a beginner and if I can do so can you!! ~Nick __ Do you Yahoo!? http://promotions.yahoo.com/new_mail __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Jose Quesada, PhD. [EMAIL PROTECTED] ESRC Postdoctoral Fellow http://lsa.colorado.edu/~quesadaj Dept. of PSychology http://www.andrew.cmu.edu/~jquesada University of Warwick office H114 Phone: +44 024 765 23 759 Coventry, UK __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Regression slope confidence interval
Hi list, is there any direct way to obtain confidence intervals for the regression slope from lm, predict.lm or the like? (If not, is there any reason? This is also missing in some other statistics softwares, and I thought this would be quite a standard application.) I know that it's easy to implement but it's for explanation to people who faint if they have to do their own programming... Christian *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Regression slope confidence interval
?confint For example: ctl - c(4.17,5.58,5.18,6.11,4.50,4.61,5.17,4.53,5.33,5.14) trt - c(4.81,4.17,4.41,3.59,5.87,3.83,6.03,4.89,4.32,4.69) group - gl(2,10,20, labels=c(Ctl,Trt)) weight - c(ctl, trt) lm(weight ~ group) Call: lm(formula = weight ~ group) Coefficients: (Intercept) groupTrt 5.032 -0.371 confint(lm(weight ~ group)) 2.5 %97.5 % (Intercept) 4.569340 5.4946602 groupTrt-1.025300 0.2833003 Christian Hennig wrote: Hi list, is there any direct way to obtain confidence intervals for the regression slope from lm, predict.lm or the like? (If not, is there any reason? This is also missing in some other statistics softwares, and I thought this would be quite a standard application.) I know that it's easy to implement but it's for explanation to people who faint if they have to do their own programming... Christian *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 452-1424 (M, W, F) fax: (917) 438-0894 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Regression slope confidence interval
?confint -Oprindelig meddelelse- Fra: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] På vegne af Christian Hennig Sendt: 29. september 2005 13:19 Til: r-help-request Mailing List Emne: [R] Regression slope confidence interval Hi list, is there any direct way to obtain confidence intervals for the regression slope from lm, predict.lm or the like? (If not, is there any reason? This is also missing in some other statistics softwares, and I thought this would be quite a standard application.) I know that it's easy to implement but it's for explanation to people who faint if they have to do their own programming... Christian *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Regression slope confidence interval
On Thu, 29 Sep 2005, Christian Hennig wrote: Hi list, is there any direct way to obtain confidence intervals for the regression slope from lm, predict.lm or the like? There is a confint method: e.g., R fm - lm(dist ~ speed, data = cars) R confint(fm, parm = speed) 2.5 % 97.5 % speed 3.096964 4.767853 hth, Z (If not, is there any reason? This is also missing in some other statistics softwares, and I thought this would be quite a standard application.) I know that it's easy to implement but it's for explanation to people who faint if they have to do their own programming... Christian *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Regression slope confidence interval
?confint Thank you to all of you. As far as I see this is not mentioned on the lm help page (though I presumably don't have the recent version), which I would suggest... Best, Christian On Thu, 29 Sep 2005, Chuck Cleland wrote: ?confint For example: ctl - c(4.17,5.58,5.18,6.11,4.50,4.61,5.17,4.53,5.33,5.14) trt - c(4.81,4.17,4.41,3.59,5.87,3.83,6.03,4.89,4.32,4.69) group - gl(2,10,20, labels=c(Ctl,Trt)) weight - c(ctl, trt) lm(weight ~ group) Call: lm(formula = weight ~ group) Coefficients: (Intercept) groupTrt 5.032 -0.371 confint(lm(weight ~ group)) 2.5 %97.5 % (Intercept) 4.569340 5.4946602 groupTrt-1.025300 0.2833003 Christian Hennig wrote: Hi list, is there any direct way to obtain confidence intervals for the regression slope from lm, predict.lm or the like? (If not, is there any reason? This is also missing in some other statistics softwares, and I thought this would be quite a standard application.) I know that it's easy to implement but it's for explanation to people who faint if they have to do their own programming... Christian *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 452-1424 (M, W, F) fax: (917) 438-0894 *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Problem with memory footprint of qq plot generated with lattice
Dave, qqmath(~val|ind,data=xx ,distribution=function(p) qt(p,df=19) ,ylab=Sample Quatinles ,xlab=Theoretical Quantiles ,aspect=1 ,prepanel = prepanel.qqmathline ,panel=function(x,y) { panel.qqmathline(y, distribution=function(p) qt(p,df=19),col=2) panel.qqmath(x, y , distribution=function(p) qt(p,df=19),pch=.,cex=2) } ) Adding f.value=fn as argument to qqmath reduces the size of the image, but neither the axis (absicissae) nor the line added by panel.qqmathline are right. Adding f.value=fn as argument to panel.qqmathline and panel.qqmath generates the right graphic, but the size of the image is again 20 MB. Any Suggestions? Eryk [EMAIL PROTECTED] wrote: nwew [EMAIL PROTECTED] wrote: Dear R helpers, I generate a qq plot using the following function call. ... dim(xx) [1] 680237 2 How about doing something like this: fn - function(n,cut=0.001,m=1000) { p - ppoints(n) p - p[pmin(p, 1-p) cut] q - pt(seq(qt(cut,df=19),qt(1-cut,df=19),length=m),df=19) sort(c(p,q)) } then adding 'f.value=fn' to your qqmath arguments? This essentially says, plot the individual data points in the extreme tails of the distribution (p 0.001 or p 0.999), and evaluate the distribution at a sparse set of points in between, where the density means you can't discern the individual values anyway. -- Dave __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Regression slope confidence interval
Why not use vcov() and the normal approximation ? ctl - c(4.17,5.58,5.18,6.11,4.50,4.61,5.17,4.53,5.33,5.14) trt - c(4.81,4.17,4.41,3.59,5.87,3.83,6.03,4.89,4.32,4.69) group - gl(2,10,20, labels=c(Ctl,Trt)) weight - c(ctl, trt) lm.D9 - lm(weight ~ group) cbind(estimate = coef(lm.D9), + lower = coef(lm.D9) - 1.96 * diag(vcov(lm.D9)), + upper = coef(lm.D9) + 1.96 * diag(vcov(lm.D9))) estimate lower upper (Intercept)5.032 4.9369482 5.1270518 groupTrt -0.371 -0.5611037 -0.1808963 To address your needs, it might also be possible to write a method for the generic of intervals() in package nlme. Best, Renaud Christian Hennig a écrit : Hi list, is there any direct way to obtain confidence intervals for the regression slope from lm, predict.lm or the like? (If not, is there any reason? This is also missing in some other statistics softwares, and I thought this would be quite a standard application.) I know that it's easy to implement but it's for explanation to people who faint if they have to do their own programming... Christian *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Renaud Lancelot Directeur Adjoint chargé des Affaires Scientifiques Deputy Director for Scientific Affairs Département EMVT du CIRAD, TA 30/B Campus International de Baillarguet 34398 Montpellier Cedex 5 - France Tel. +33 (0)4 67 59 37 17 Secr. +33 (0)4 67 59 39 04 Fax +33 (0)4 67 59 37 95 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Regression slope confidence interval
Sorry, I forgot confint and I made a mistake in my suggestion which should be: cbind(estimate = coef(lm.D9), lower = coef(lm.D9) - 1.96 * sqrt(diag(vcov(lm.D9))), upper = coef(lm.D9) + 1.96 * sqrt(diag(vcov(lm.D9 Best, Renaud Christian Hennig a écrit : Hi list, is there any direct way to obtain confidence intervals for the regression slope from lm, predict.lm or the like? (If not, is there any reason? This is also missing in some other statistics softwares, and I thought this would be quite a standard application.) I know that it's easy to implement but it's for explanation to people who faint if they have to do their own programming... Christian *** --- *** Christian Hennig University College London, Department of Statistical Science Gower St., London WC1E 6BT, phone +44 207 679 1698 [EMAIL PROTECTED], www.homepages.ucl.ac.uk/~ucakche __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Renaud Lancelot Directeur Adjoint chargé des Affaires Scientifiques Deputy Director for Scientific Affairs Département EMVT du CIRAD, TA 30/B Campus International de Baillarguet 34398 Montpellier Cedex 5 - France Tel. +33 (0)4 67 59 37 17 Secr. +33 (0)4 67 59 39 04 Fax +33 (0)4 67 59 37 95 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] memory issues with large data set
memory.limit may not be the correct command. I use the command 'utils:: memory.size(3*1024)' to increase my memory size after using editbin to modify the header of R to make it LARGEADDRESSAWARE as described in the above FAQ. I am able to read about 2.7Gb into memory that way with 4Gb of ram. Not only am I able to read it into memory, but I can do regessions on subsets of the data no problem. My question has always been, why can't R ship LARGEADDRESSAWARE for those users who may not have access to 'editbin' type tools? Thanks, Roger On 9/28/05, Christina Yau [EMAIL PROTECTED] wrote: Hi, I am running R 2.0.1.1 http://2.0.1.1. on Windows. It is a Dell Dimension with a 3.2 Ghz Processor and 4Gb RAM. When using the ReadAffy() function to read in 97 arrays, I get the below error messages: Error: cannot allocate vector of size 393529 Reached total allocation of 1024Mb: see help(memory.size) When I use the comman memory.limit(size=4000) to increase the memory size to the maximum available, I got a NULL as a response. I proceeded to re-run ReadAffy(). This time, I only get the first error message. Error: cannot allocate vector of size 393529 From what I've read, this is more of a problem with Windows than with R. But I am wondering if there is anything I can do, either with the set up of R or Windows, to solve this problem and read the data set into R using this machine. Thank you for your attention, Christina __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] anova on binomial LMER objects
On Wed, 28 Sep 2005, Robert Bagchi wrote: Hi Patrick thanks for your advice. I have now tried glmmPQL, and it worked fine - I'm getting consistent results between plots and models fitted by glmmPQL. Plus it allows predict() and resid() which is another advantage over lmer at present. quick question though: why does one need to use PQL for binomial models? Is there a good reference for this? You don't have to use PQL for binomial models, but you can't use least-squares. PQL is an approximate solution. Laplace and Adaptive Gaussian Quadrature options in lmer are better approximations. So lmer would likely become the better option as it progresses in its development (though the current issues you've found with the F ratios certainly sound like maybe lmer isn't better for you in its current incarnation). alan -- Alan B. Cobo-Lewis, Ph.D. (207) 581-3840 tel Department of Psychology(207) 581-6128 fax University of Maine Orono, ME 04469-5742[EMAIL PROTECTED] http://www.umaine.edu/visualperception __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Easy cut paste from Excel to R?
See: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/26922.html On 9/29/05, Jose Quesada [EMAIL PROTECTED] wrote: Sorry to revive and old topic, but writing to the clipboard seems to have a problem for me: column names are ignored. Example: # ~~~ # write.clipboard # ~~~ write.clipboard = function(obj) { write.table(obj, file(clipboard), sep=\t, row.names=F, col.names=T) } a= matrix(1:4,2,2) colnames(a) = c(a, b) write.clipboard(a) a = as.data.frame(a) write.clipboard(a) both attempts will paste the date without column names. Any idea why? Thanks, -Jose On 2/16/05, Werner Wernersen [EMAIL PROTECTED] wrote: Thank you all very much for the answers! The read.table / read.delim2 commands are exactly what I was looking for to get a couple of numbers or a little matrix quickly into R without creating an extra text file every time. And it works the other way around as well: write.table(x, file(clipboard), sep=\t) Fantastic! Thanks again, Werner Nick Drew wrote: I've had good luck with the scan() function when I want to get a few numbers from Excel into R quickly to use it as a calculator. CAVEAT: you have to have the numbers you want to copy in a column not a row in Excel. For example: In Excel your data are in a column as follows: Col A 1 2 3 Then copy the 3 cells (e.g. 1, 2,3) in Excel and open R and type in: data - scan() Then Paste using Ctrl-V. Hit the Enter key. You know have an object called data that you can use and manipulate in R. I've taken this even further by creating an R function that will take a column of numbers from Excel and then scan() them into R, create a matrix, and then perform a Chi-square test. Let me know if you'd like to know more. I'm a beginner and if I can do so can you!! ~Nick __ Do you Yahoo!? http://promotions.yahoo.com/new_mail __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Jose Quesada, PhD. [EMAIL PROTECTED] ESRC Postdoctoral Fellow http://lsa.colorado.edu/~quesadaj Dept. of PSychology http://www.andrew.cmu.edu/~jquesada University of Warwick office H114 Phone: +44 024 765 23 759 Coventry, UK __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] standard error of variances and covariances of the random effects with LME
Hello, how do I obtain standard errors of variances and covariances of the random effects with LME comparable to those of for example MlWin? I know you shouldn't use them because the distribution of the estimator isn't symmetric blablabla, but I need a measure of the variance of those estimates for pooling my multiple imputation results. Regards, Roel. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] reshaping data?
I have a file like this: a 0.1 a 0.2 a 0.9 b 0.5 b 0.9 b 0.7 c 0.6 c 0.99 c 0.88 Which I would like to get to be the following matrix: 0.1 0.20.30.4 ... a 12 0 0 b 00 0 0 .. I.e: each place in the matrix denotes how many entries in each category that are betwee 0.0 and 0.1, 0.1 and 0.2 and so on. The way I was thinking of doing it was by constructing an empty matrix and then doing a for loop testing each element and incrementing in the matrix as appropriate. However, it struck me that this has to be easier to do than that. Am I right? Karin -- Karin Lagesen, PhD student [EMAIL PROTECTED] http://www.cmbn.no/rognes/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] standard error of variances and covariances of the random effects with LME
You cannot. Also, it's not that the distribution of the random effects is not symmetric, but that it *may* not be symmetric, and this is an assumption that should be checked. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roel de Jong Sent: Thursday, September 29, 2005 9:20 AM To: r-help@stat.math.ethz.ch Subject: [R] standard error of variances and covariances of the random effects with LME Hello, how do I obtain standard errors of variances and covariances of the random effects with LME comparable to those of for example MlWin? I know you shouldn't use them because the distribution of the estimator isn't symmetric blablabla, but I need a measure of the variance of those estimates for pooling my multiple imputation results. Regards, Roel. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] boxplot and xlim confusion?
Petr Pikal [EMAIL PROTECTED] writes: Hi Karin I did not have seen any answer for your question yet so here is a try. I gues you want the horizotal layout or your boxplot. boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=letters[1:3]) boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=c(NA,b,NA)) These look like the plots I want, yes. However, is there a way of locking the x-axis? When I do these several times in a row, the range of the x axis moves with the data being plotted. If I set xlim in the boxplot command, it makes no difference what so ever. So this is closest what I could dig from your text. ***reproducible*** example would by good starting point what you want and was not able to produce. Sorry about that...:) Karin -- Karin Lagesen, PhD student [EMAIL PROTECTED] http://www.cmbn.no/rognes/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] reshaping data?
you could use something like this (but maybe there are better proposals): dat - data.frame(g = rep(letters[1:3], each = 5), val = runif(15)) out - do.call(rbind, lapply(split(dat$val, dat$g), function(x){ f - factor(findInterval(x, vec = seq(0, 1, 0.1)), levels = 1:10) table(f) })) colnames(out) - seq(0.1, 1, 0.1) out I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Karin Lagesen [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Thursday, September 29, 2005 3:16 PM Subject: [R] reshaping data? I have a file like this: a 0.1 a 0.2 a 0.9 b 0.5 b 0.9 b 0.7 c 0.6 c 0.99 c 0.88 Which I would like to get to be the following matrix: 0.1 0.20.30.4 ... a 12 0 0 b 00 0 0 .. I.e: each place in the matrix denotes how many entries in each category that are betwee 0.0 and 0.1, 0.1 and 0.2 and so on. The way I was thinking of doing it was by constructing an empty matrix and then doing a for loop testing each element and incrementing in the matrix as appropriate. However, it struck me that this has to be easier to do than that. Am I right? Karin -- Karin Lagesen, PhD student [EMAIL PROTECTED] http://www.cmbn.no/rognes/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] reshaping data?
On Thu, 29 Sep 2005, Karin Lagesen wrote: I have a file like this: a 0.1 a 0.2 a 0.9 b 0.5 b 0.9 b 0.7 c 0.6 c 0.99 c 0.88 Which I would like to get to be the following matrix: 0.1 0.20.30.4 ... a 12 0 0 b 00 0 0 .. I.e: each place in the matrix denotes how many entries in each category that are betwee 0.0 and 0.1, 0.1 and 0.2 and so on. If your variables are called, say, id and value, and value is between 0 and 1 then table(id,cut(value,breaks=(0:10)/10)) works. -thomas __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] New family for gam in the mgcv library
Hi! I'm using R 2.0.1 on a Sun, with mgcv library version 1.3-1. I would like to implement a new family for the function gam in mgcv (truncated Poisson family as defined in Barry Welsh (2002), Ecological Modelling). I therefore defined a family function with all the necessary components (linkfun, linkinv, variance, etc). But then I run into problems because my link function is not a standard one, and fix.family.link() and fix.family.var() (called by gam.outer) won't work. As I understand, the purpose of these two functions is to modify families so to add a dvar component and a d2link component, necessary for the fitting. The fix I can imagine (but I didn't try yet) is to add the two extra components (dvar and d2link) when defining the new family, and to redefine the function gam.outer to avoid the use of fix.family.*. Does someone have experience with this? Would this be the end of my troubles? Thank you for your help, Eva Cantoni -- Dr Eva Cantoni phone : (+41) 22 379 8240 Econométrie - Univ. Genève fax: (+41) 22 379 8299 40, Bd du Pont d'Arve e-mail : [EMAIL PROTECTED] CH-1211 Genève 4 http://www.unige.ch/ses/metri/cantoni __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] reshaping data?
This can be shortened slightly using cut: table(data.frame(g = dat$g, val = cut(dat$val, 0:10/10, lab = 1:10/10))) On 9/29/05, Dimitris Rizopoulos [EMAIL PROTECTED] wrote: you could use something like this (but maybe there are better proposals): dat - data.frame(g = rep(letters[1:3], each = 5), val = runif(15)) out - do.call(rbind, lapply(split(dat$val, dat$g), function(x){ f - factor(findInterval(x, vec = seq(0, 1, 0.1)), levels = 1:10) table(f) })) colnames(out) - seq(0.1, 1, 0.1) out I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Karin Lagesen [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Thursday, September 29, 2005 3:16 PM Subject: [R] reshaping data? I have a file like this: a 0.1 a 0.2 a 0.9 b 0.5 b 0.9 b 0.7 c 0.6 c 0.99 c 0.88 Which I would like to get to be the following matrix: 0.1 0.20.30.4 ... a 12 0 0 b 00 0 0 .. I.e: each place in the matrix denotes how many entries in each category that are betwee 0.0 and 0.1, 0.1 and 0.2 and so on. The way I was thinking of doing it was by constructing an empty matrix and then doing a for loop testing each element and incrementing in the matrix as appropriate. However, it struck me that this has to be easier to do than that. Am I right? Karin -- Karin Lagesen, PhD student [EMAIL PROTECTED] http://www.cmbn.no/rognes/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] reshaping data?
And slightly more to: table(g = dat$g, val = cut(dat$val, 0:10/10, lab = 1:10/10)) On 9/29/05, Gabor Grothendieck [EMAIL PROTECTED] wrote: This can be shortened slightly using cut: table(data.frame(g = dat$g, val = cut(dat$val, 0:10/10, lab = 1:10/10))) On 9/29/05, Dimitris Rizopoulos [EMAIL PROTECTED] wrote: you could use something like this (but maybe there are better proposals): dat - data.frame(g = rep(letters[1:3], each = 5), val = runif(15)) out - do.call(rbind, lapply(split(dat$val, dat$g), function(x){ f - factor(findInterval(x, vec = seq(0, 1, 0.1)), levels = 1:10) table(f) })) colnames(out) - seq(0.1, 1, 0.1) out I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Karin Lagesen [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Thursday, September 29, 2005 3:16 PM Subject: [R] reshaping data? I have a file like this: a 0.1 a 0.2 a 0.9 b 0.5 b 0.9 b 0.7 c 0.6 c 0.99 c 0.88 Which I would like to get to be the following matrix: 0.1 0.20.30.4 ... a 12 0 0 b 00 0 0 .. I.e: each place in the matrix denotes how many entries in each category that are betwee 0.0 and 0.1, 0.1 and 0.2 and so on. The way I was thinking of doing it was by constructing an empty matrix and then doing a for loop testing each element and incrementing in the matrix as appropriate. However, it struck me that this has to be easier to do than that. Am I right? Karin -- Karin Lagesen, PhD student [EMAIL PROTECTED] http://www.cmbn.no/rognes/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] boxplot and xlim confusion?
On Thu, 2005-09-29 at 15:28 +0200, Karin Lagesen wrote: Petr Pikal [EMAIL PROTECTED] writes: Hi Karin I did not have seen any answer for your question yet so here is a try. I gues you want the horizotal layout or your boxplot. boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=letters[1:3]) boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=c(NA,b,NA)) These look like the plots I want, yes. However, is there a way of locking the x-axis? When I do these several times in a row, the range of the x axis moves with the data being plotted. If I set xlim in the boxplot command, it makes no difference what so ever. That's because when you use 'horizontal = TRUE' in boxplot(), the x and y axes are rotated. Thus you need to set 'ylim' and not 'xlim', where ylim is the range of values in your data irrespective of the axis orientation. This is done in bxp(), which is the actual function doing the plotting for boxplot(). In bxp(), there is the follow code snippet: if (horizontal) plot.window(ylim = c(0.5, n + 0.5), xlim = ylim, log = log) else plot.window(xlim = c(0.5, n + 0.5), ylim = ylim, log = log) Note in the first case, that the xlim value for plot.window() is set to the ylim value passed from boxplot() or from bxp(), if called directly. So, the above examples by Petr should be something like: boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal = TRUE, names=letters[1:3], ylim = c(-4, 4)) boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =TRUE, names=c(NA,b,NA), ylim = c(-4, 4)) HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R-2.1.1 on AIX 5.*
Hi, I need to install R-2.1.1 on a AIX machine. (i have tried several machine with different os version : 5.1, 5.2 and 5.3) I have tried several configuration which was advise on the R-install and admin section : but none goes to the end of compilation. my configure command line is : #./configure --prefix=$HOME/local/R-2.1.1 CC=xlc_r CXX=xlC_r F77=xlf_r --without-x --without-readline OBJECT_MODE=64 LDFLAGS=-brtl CFLAGS=-O -qstrict FFLAGS=-O -qstrict CXXFLAGS=-O -qstrict then i compiled the source : #make and i have the following error. making vsnprintf.d from vsnprintf.c xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c Rmain.c -o Rmain.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c CConverters.c -o CConverters.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c CommandLineArgs.c -o CommandLin eArgs.o CommandLineArgs.c, line 170.32: 1506-1298 (W) The subscript 31 is out of range. The valid range is 0 to 30. xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c Rdynload.c -o Rdynload.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c Renviron.c -o Renviron.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c RNG.c -o RNG.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c apply.c -o apply.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c arithmetic.c -o arithmetic.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c apse.c -o apse.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c array.c -o array.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c attrib.c -o attrib.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c base.c -o base.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c bind.c -o bind.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c builtin.c -o builtin.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c character.c -o character.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c coerce.c -o coerce.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c colors.c -o colors.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c complex.c -o complex.o xlc_r -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/pcre -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -O -qstrict -c connections.c -o connections.o connections.c, line 2580.18: 1506-052 (S) Duplicate case label for value 4. Labels must be unique. connections.c, line 2598.18: 1506-052 (S) Duplicate case label for value 4. Labels must be unique. connections.c, line 2757.18: 1506-052 (S) Duplicate case label for value 4. Labels must be unique. connections.c, line
[R] Repeated measure Generalized Linear Mixed Model (glmm) ?
Dear all, I want to know the probability of dying in a trap as a function of habitat variables for a metapopulation of voles sampled in 44 stations. My dataset is as follow: Station tag number deadhabitat1habitat2 1 1 yes 20 26 1 2 no 20 26 2 3 no 15 16 .. As far as I know, I should use a mixed-model as: glmmPQL-(fixed = dead~habitat1+habitat2, random = ~1|station, family = binomial) in the MASS library. However, some vole individuals were recaptured (but it is not a study designed for capture-mark-recapture!!!) therefore, I have for example, the 4th vole: Station tag number deadhabitat1habitat2 3 4 no 10 12 3 4 yes 10 12 Someone suggested me to use glmm with repeated measures with the number of tag as the variable to repeat. Apparently, it is possible to do this in SAS. QUESTIONS : 1) Is this also possible to do this in R, and if yes, could you please tell me how? 2) Is this a valid way to analyse this data, and if not, could someone please put me in the right way? Thank you very very much to all in advance Jérôme Lemaître Étudiant au doctorat Université Laval Québec, QC G1K 7P4 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Cox regression on interval censoring
Regarding non-parametric regression with interval censored survival data: Does anyone know where to find an extension dealing with Cox regression for (overlapping) interval censored data? If not, maybe someone has some ML expressions for this interval scenario laying around that you are willing to share? I would be really really greatful! Thanks and best regards // David * D. Hannersjoe Department of Aquaculture Swedish University of Agricultural Science SWEDEN E-mail: [EMAIL PROTECTED] * ### This message has been scanned by F-Secure Anti-Virus for Mic...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Fast AUC computation
See colAUC in caTools (there is a problem with 1.3 version, 1.4 is on the way). See examples for other functions calculating AUC. An alternative approach is: x1 = x[y==1]; n1 = length(x1); x2 = x[y==0]; n2 = length(x2); r = rank(c(x1,x2)) auc = (sum(r[1:n1]) - n1*(n1+1)/2) / (n1*n2) Which is very fast. Jarek \ Jarek Tuszynski, PhD. o / \ Science Applications International Corporation \__,| (703) 676-4192 \ [EMAIL PROTECTED] ` \ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Nina Paynter Sent: Wednesday, September 28, 2005 4:43 PM To: r-help@stat.math.ethz.ch Subject: [R] Fast AUC computation I am doing a simulation with a relatively large data set (20,000 observations) for which I want to calculate the area under the Receiver Operator Curve (AUC) for many parameter combinations. I am using the ROC library and the following commands to generate each AUC: rocobj=rocdemo.sca(truth = ymis, data = model$fitted.values, rule = dxrule.sca) #generation of observed ROC object aucobj=AUC(rocobj) #pulling out just the observed AUC - trapezoidal not integrated but they are pretty slow. Does anyone know of a faster way to get the AUC? Thanks, Nina [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Fisher's discriminant functions
Hi everyone, I'm trying to solve a problem about how to get the Fisher's discriminant functions of a lda (linear discriminant analysis) object, I mean, the object obtained from doing lda(formula, data) function of the package MASS in R-project. This object gives me the canonical linear functions (n-1 coefficients matrix of n groups at least), and only with this information I could predict the group of an observation data using the predict function. But what I need is the Fisher's discriminant functions (n coefficients matrix of n groups) in order to classify my future data. The object predict gives me only the following attributes x, posterior and class, but none of them are the coefficients matrix of the Fisher's discriminant functions, and the reason why I'm not using the predict function for my predictions is because the time spent is very high for what I'm expecting, about 0.5 seconds while I can obtain this prediction with the Fisher's discriminant functions faster. So, I don't know if there's a package which I can use to obtain the mentioned coefficients matrix of the Fisher's discriminant functions. I anyone can help, I would appreciate it greatly. Thank you and regards. Carlos Niharra López __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Error using a data frame as the start parameter in mle()
Dear R-Users, I am trying to use mle() to optimize two (or more) parameters, but I want to specify those parmeters in a data frame rather than having to spell them out separately in the start variable of mle(). My call is mle(negll, start=list(aps=init), fixed=list(measphot=newphot, formod=formod, Nbands=Nbands), method=BFGS) where negll is a function I have written which uses the function predict.loess(). negll works fine when called directly. The parameter I am trying to optimize, aps, is a data frame containing two parameters, e.g. init teff logg 1 8000 4.5 When I run mle I get the following error message Error in predict.loess(formod[[band]], aps) : Argument aps is missing, with no default As negll does work fine, I presume I am incorrectly passing aps into mle(). Note that mle() works fine if I rewrite negll to work on a scalar aps and then I use start=list(aps=500), for example. Can anyone help me with this? Incidentally, I am only using a data frame for aps because I am using loess(), and this seems to require a formula with named variables in a data frame (here logg and teff). I can't get it work with arrays: temp - loess(formula = photd[, band] ~ gridaps[, 1] * gridaps[, 2]) predict(temp, c(4,8000)) Error in predict.loess(temp, c(4, 8000)) : newdata does not contain the variables needed Thanks in advance for any clues. Coryn. --- Coryn Bailer-Jones[EMAIL PROTECTED] Max-Planck-Institut fuer Astronomie http://www.mpia-hd.mpg.de/homes/calj/ Koenigstuhl 17tel: +49 6221 528-224(direct) D-69117 Heidelberg +49 6221 528-0 (reception) Germany fax: +49 6221 528-246 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Easy cut paste from Excel to R?
Hi Works for me. write.excel - function (tab, ...) write.table(tab, clipboard, sep = \t, row.names = F) write.excel(a) from your example shows in Excel after ctrl-V as a table with names. HTH Petr On 29 Sep 2005 at 12:12, Jose Quesada wrote: Date sent: Thu, 29 Sep 2005 12:12:00 +0100 From: Jose Quesada [EMAIL PROTECTED] To: Werner Wernersen [EMAIL PROTECTED] Copies to: r-help@stat.math.ethz.ch Subject:Re: [R] Easy cut paste from Excel to R? Send reply to: Jose Quesada [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] mailto:[EMAIL PROTECTED] Sorry to revive and old topic, but writing to the clipboard seems to have a problem for me: column names are ignored. Example: # ~~~ # write.clipboard # ~~~ write.clipboard = function(obj) { write.table(obj, file(clipboard), sep=\t, row.names=F, col.names=T) } a= matrix(1:4,2,2) colnames(a) = c(a, b) write.clipboard(a) a = as.data.frame(a) write.clipboard(a) both attempts will paste the date without column names. Any idea why? Thanks, -Jose On 2/16/05, Werner Wernersen [EMAIL PROTECTED] wrote: Thank you all very much for the answers! The read.table / read.delim2 commands are exactly what I was looking for to get a couple of numbers or a little matrix quickly into R without creating an extra text file every time. And it works the other way around as well: write.table(x, file(clipboard), sep=\t) Fantastic! Thanks again, Werner Nick Drew wrote: I've had good luck with the scan() function when I want to get a few numbers from Excel into R quickly to use it as a calculator. CAVEAT: you have to have the numbers you want to copy in a column not a row in Excel. For example: In Excel your data are in a column as follows: Col A 1 2 3 Then copy the 3 cells (e.g. 1, 2,3) in Excel and open R and type in: data - scan() Then Paste using Ctrl-V. Hit the Enter key. You know have an object called data that you can use and manipulate in R. I've taken this even further by creating an R function that will take a column of numbers from Excel and then scan() them into R, create a matrix, and then perform a Chi-square test. Let me know if you'd like to know more. I'm a beginner and if I can do so can you!! ~Nick __ Do you Yahoo!? http://promotions.yahoo.com/new_mail __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Jose Quesada, PhD. [EMAIL PROTECTED] ESRC Postdoctoral Fellow http://lsa.colorado.edu/~quesadaj Dept. of PSychology http://www.andrew.cmu.edu/~jquesada University of Warwick office H114 Phone: +44 024 765 23 759 Coventry, UK __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] boxplot and xlim confusion?
Hi So you need x to be same in all subsequent plots e.g. from -5 to 5 like in this. boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=letters[1:3], ylim=c(-5,5)) You find it in a thorough reading bxp man page. Currently, 'ylim' is used 'along the boxplot', i.e., vertically, when 'horizontal' is false. So if horizontal is true you still need to use ylim not xlim :-) HTH Petr On 29 Sep 2005 at 15:28, Karin Lagesen wrote: To: Petr Pikal [EMAIL PROTECTED] From: Karin Lagesen [EMAIL PROTECTED] Date sent: Thu, 29 Sep 2005 15:28:14 +0200 Copies to: [EMAIL PROTECTED], Karin Lagesen [EMAIL PROTECTED] Subject:Re: [R] boxplot and xlim confusion? Petr Pikal [EMAIL PROTECTED] writes: Hi Karin I did not have seen any answer for your question yet so here is a try. I gues you want the horizotal layout or your boxplot. boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=letters[1:3]) boxplot(split(rnorm(30), rep(1:3, each=10)), horizontal =T, names=c(NA,b,NA)) These look like the plots I want, yes. However, is there a way of locking the x-axis? When I do these several times in a row, the range of the x axis moves with the data being plotted. If I set xlim in the boxplot command, it makes no difference what so ever. So this is closest what I could dig from your text. ***reproducible*** example would by good starting point what you want and was not able to produce. Sorry about that...:) Karin -- Karin Lagesen, PhD student [EMAIL PROTECTED] http://www.cmbn.no/rognes/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] quasi-random vector according to an independent graph
Are you still interested in a reply to this post? I have not seen any. If you are, it might help if you were more specific, e.g., following the posting guide www.R-project.org/posting-guide.html. I'm not certain what you mean by a joint distribution defined by an independent graph, and my efforts using RSiteSearch exposed several things that might be useful but none that seemed to me to be obvious answers to your question. Sorry I could not be more helpful. spencer graves Jinfang Wang wrote: Dear R-users, Is anyone aware of any function/package for generating a random vector from a joint distribution defined by an independent graph? Or I have to work it out myself? Thanks. Jinfang -- Jinfang Wang, Associate Professor Chiba University, Japan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to get the rowindices without using which?
I'm not certain what you are asking. Consider the following: set.seed(1) (irows - sample(1:nrow(iris), 10)) iris[irows,] If you want more than this, PLEASE do read the posting guide! www.R-project.org/posting-guide.html. I believe that people who follow the posting guide generally get quicker and better answers than those who don't. spencer graves Martin Lam wrote: Hi, I was wondering if it is possible to get the rowindices without using the function which because I don't have a restriction criteria. Here's an example of what I mean: # take 10 randomly selected instances iris[sample(1:nrow(iris), 10),] # output Sepal.Length Sepal.Width Petal.Length Petal.Width Species 76 6.6 3.0 4.4 1.4 versicolor 105 6.5 3.0 5.8 2.2 virginica 131 7.4 2.8 6.1 1.9 virginica 79 6.0 2.9 4.5 1.5 versicolor 69 6.2 2.2 4.5 1.5 versicolor 42 4.5 2.3 1.3 0.3 setosa 25 4.8 3.4 1.9 0.2 setosa 129 6.4 2.8 5.6 2.1 virginica 60 5.2 2.7 3.9 1.4 versicolor 80 5.7 2.6 3.5 1.0 versicolor What I want to get are their rownumbers: 76, 105, 131, 79, 69, 42, 25, 129, 60, 80. Thanks in advance, 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 -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Producing empirical bayes estimates in disease mapping for lognormal model
I just got 74 hits from RSiteSearch(empirical bayes), 0 from RSiteSearch(empirical bayes lognormal), and 18 from RSiteSearch(empirical bayes normal). Have you tried this? If you still would like information from this list, PLEASE do read the posting guide! www.R-project.org/posting-guide.html. I believe that people who do usually get better answers quicker. spencer graves Oarabile Molaodi wrote: I'm trying to produce empirical bayes estimates based on the lognormal model in disease mapping Is there a way this can be done in R? thanks Oarabile __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] priceIts
Prof Brian Ripley ripley at stats.ox.ac.uk writes: chart.yahoo.com is notoriously fickle (as we have seen from the tseries Maybe we should update the base URL. I maintain CPAN's Finance::YahooQuote (and am author of two apps, also on CPAN, which use it fairly heavily) and this has been nothing but rock solid in the six years that I've been scraping data from Yahoo, often at a once-a-minute frequency (for the smtm quote display tool). Of note is that Finance::YahooQuote uses a different machine: $QURLbase = http://quote.yahoo.com/d?f=;; but both chart.yahoo.com and quote.yahoo.com appear to be on the same subnet. Not sure I'll have time to patch tseries, but it is something we may need to keep in mind. Hth, Dirk __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Display values in piechart/barplot
Is it possible to automatically display the underlying values of a piechart/barplot in the graphic? If so, which package/function/argument do I need for it? Thanks, Volker __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] How to add frame (frame.plot=T) to Plot.Design?
Hi R-help, When using the package Design and the plot.Design function all the graphs of an lrm.fit (lrm()) are plotted without a frame (only with axes). How can the frame be added to these plots? In the plot.default function the frame can be added/or removed via the frame.plot=T/F, respectively. e.g., plot(1,axes=T,frame.plot=T). How can this be done with plot.Design(lrm.fit) Thanks, Jan ___ Ir. Jan Verbesselt Research Associate Biosystems Department ~ M³-BIORES Vital Decosterstraat 102, 3000 Leuven, Belgium Tel: +32-16-329750 Fax: +32-16-329760 http://gloveg.kuleuven.ac.be/ ___ Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] regsubsets selection criterion
Questions like this are best directed to the package maintainer(s). From help(package=leaps), I learned that Thomas Lumley is the author and maintainer for leaps; I'm including him as a 'cc', so he can correct or add to my comments if he feels so inclined. After some searching, I learned that ?plot.regsubsets includes an argument scale=c(bic, Cp, adjr2, r2). From this, I infer that you have your choice of these four criteria. hope this helps. spencer graves Samuel Bertrand wrote: Hello, I am using the 'regsubsets' function (from leaps package) to get the best linear models to explain 1 variable from 1 to 5 explanatory variables (exhaustive search). Is there anyone who can tell me on which criterion is based the 'regsubsets' function ? Thank you. samuel Samuel BERTRAND Doctorant Laboratoire de Biomecanique LBM - ENSAM - CNRS UMR 8005 http://bio-web.paris.ensam.fr 151, bd de l'Hopital 75013 PARIS Tel. +33 (0) 1 44 24 64 53 Fax +33 (0) 1 44 24 63 66 [[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 -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] regsubsets selection criterion
On Thu, 29 Sep 2005, Spencer Graves wrote: Questions like this are best directed to the package maintainer(s). From help(package=leaps), I learned that Thomas Lumley is the author and maintainer for leaps; I'm including him as a 'cc', so he can correct or add to my comments if he feels so inclined. He has already answered this question when it was originally posted. -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] quasi-random vector according to an independent graph
Might the graphical models in R packages be of interest? http://www.r-project.org/gR/ Reid Huntsinger -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Spencer Graves Sent: Thursday, September 29, 2005 11:43 AM To: Jinfang Wang Cc: r-help@stat.math.ethz.ch Subject: Re: [R] quasi-random vector according to an independent graph Are you still interested in a reply to this post? I have not seen any. If you are, it might help if you were more specific, e.g., following the posting guide www.R-project.org/posting-guide.html. I'm not certain what you mean by a joint distribution defined by an independent graph, and my efforts using RSiteSearch exposed several things that might be useful but none that seemed to me to be obvious answers to your question. Sorry I could not be more helpful. spencer graves Jinfang Wang wrote: Dear R-users, Is anyone aware of any function/package for generating a random vector from a joint distribution defined by an independent graph? Or I have to work it out myself? Thanks. Jinfang -- Jinfang Wang, Associate Professor Chiba University, Japan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Display values in piechart/barplot
On Thu, 2005-09-29 at 14:34 +0200, Volker Rehbock wrote: Is it possible to automatically display the underlying values of a piechart/barplot in the graphic? If so, which package/function/argument do I need for it? Thanks, Volker Using pie charts are not a particularly good way of displaying data, even though the pie() function is available in R. See ?pie for more information on this. For barplots, the following provide two approaches: # Place the bar values above the bars # Note that I set 'ylim' to make room for # the text labels above the bars vals - 1:5 names(vals) - LETTERS[1:5] mp - barplot(vals, ylim = c(0, 6)) text(mp, vals, labels = vals, pos = 3) # Place the bar values below the x axis vals - 1:5 names(vals) - LETTERS[1:5] mp - barplot(vals) mtext(side = 1, at = mp, text = vals, line = 3) Note that barplot() returns the bar midpoints, so you can use these values for annotation placement. See ?barplot, ?text and ?mtext for more information. HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] plot.augPred sorted and labelled according second factor
Hi using this code example: library(nlme) fm1 - lme(Orthodont, random = ~1) plot(augPred(fm1)) is there any way to have the plots in each cell labelled and ordered according to Orthodont$Sex? I.e. in addition to the bar with the label for Orthodont$Subject there is another bar labelling the Sex of the subject? thanks a lot christoph -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] lmer random effect model matrix question
I have one fixed effect, sor, with two levels. I have eight lots and three wafers from each lot. I have included the data below. I would like to fit a mixed model that estimates a covariance parameter for wafer, which is nested in lot, and two covariance parameters for lot, one for each level of sor. The following command fits the model that I want, except for it estimates the correlation between the two covariance parameters for lot. Is there anyway to make R not estimate this correlation? Thank you. lmer(y~sor+(sor-1|lot)+(1|wafer:lot),wafer) For those familiar with proc mixed the following SAS code fits the model that I want: proc mixed scoring=4; class sor lot wafer site; model y= sor/ddfm=satterth; random lot(sor)/group=sor; random wafer(lot); run; sor lot wafer sitey 11 1 11 2006 21 1 12 1999 31 1 13 2007 41 1 21 1980 51 1 22 1988 61 1 23 1982 71 1 31 2000 81 1 32 1998 91 1 33 2007 10 1 2 11 1991 11 1 2 12 1990 12 1 2 13 1988 13 1 2 21 1987 14 1 2 22 1989 15 1 2 23 1988 16 1 2 31 1985 17 1 2 32 1983 18 1 2 33 1989 19 1 3 11 2000 20 1 3 12 2004 21 1 3 13 2004 22 1 3 21 2001 23 1 3 22 1996 24 1 3 23 2004 25 1 3 31 1999 26 1 3 32 2000 27 1 3 33 2002 28 1 4 11 1997 29 1 4 12 1994 30 1 4 13 1996 31 1 4 21 1996 32 1 4 22 2000 33 1 4 23 2002 34 1 4 31 1987 35 1 4 32 1990 36 1 4 33 1995 37 2 5 11 2013 38 2 5 12 2004 39 2 5 13 2009 40 2 5 21 2023 41 2 5 22 2018 42 2 5 23 2010 43 2 5 31 2020 44 2 5 32 2023 45 2 5 33 2015 46 2 6 11 2032 47 2 6 12 2036 48 2 6 13 2030 49 2 6 21 2018 50 2 6 22 2022 51 2 6 23 2026 52 2 6 31 2009 53 2 6 32 2010 54 2 6 33 2011 55 2 7 11 1984 56 2 7 12 1993 57 2 7 13 1993 58 2 7 21 1992 59 2 7 22 1992 60 2 7 23 1990 61 2 7 31 1996 62 2 7 32 1993 63 2 7 33 1987 64 2 8 11 1996 65 2 8 12 1989 66 2 8 13 1996 67 2 8 21 1997 68 2 8 22 1993 69 2 8 23 1996 70 2 8 31 1990 71 2 8 32 1989 72 2 8 33 1992 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to add frame (frame.plot=T) to Plot.Design?
On Thu, 2005-09-29 at 18:36 +0200, Jan Verbesselt wrote: Hi R-help, When using the package Design and the plot.Design function all the graphs of an lrm.fit (lrm()) are plotted without a frame (only with axes). How can the frame be added to these plots? In the plot.default function the frame can be added/or removed via the frame.plot=T/F, respectively. e.g., plot(1,axes=T,frame.plot=T). How can this be done with plot.Design(lrm.fit) Thanks, Jan See ?box HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Reading Health Level Seven (HL 7) data format into R
Has anyone developed code for reading HL 7 formatted data into R? HL 7 is an open standard for the digital transfer of health care information. It's moving towards XML but it isn't there yet. A study I started working on today will be receiving lab data in that format. I'd rather not reinvent any more wheels than strictly necessary. Thanks for your time. Chuck __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] need suggestion about building formual
I'm not certain what you are asking. You can build expressions in R as character strings and then execute them. Example: expr - paste(two -, 1, +, 1) eval(parse(text=expr)) two If this does not answer your question, PLEASE do read the posting guide, www.R-project.org/posting-guide.html. It can help increase the chances of a quick and useful reply. spencer graves Simple wrote: hi, I'm an newbie for R,I want do some fitting in R. I wander if it is possible to write a few of equations but only one formual when fitting Currently,My problem is,in R, is there methods combination a few equations into one formual? For example, y=f1(k); k=f2(t); t=f3(x); although it is certain that the can be equations turn into one formual as y~f(x),but write such a complexity string make me painful. I have searched the web and found out there were only examples with one formual.any suggestion? I hope that I have omit something. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] standard error of variances and covariances of the random effects with LME
With lme you could but it would take a while to work it out. There is an approximate Hessian matrix for the parameters that determine the variance-covariance matrix in there somewhere and if you were sufficiently persistent you could convert that apVar component to the scale of the variances and covariances. I believe it is in the scale of the logarithm of the standard deviation and Fisher's z transformation (i.e. the hyperbolic arc tangent) of the correlation. On 9/29/05, Doran, Harold [EMAIL PROTECTED] wrote: You cannot. Also, it's not that the distribution of the random effects is not symmetric, but that it *may* not be symmetric, and this is an assumption that should be checked. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roel de Jong Sent: Thursday, September 29, 2005 9:20 AM To: r-help@stat.math.ethz.ch Subject: [R] standard error of variances and covariances of the random effects with LME Hello, how do I obtain standard errors of variances and covariances of the random effects with LME comparable to those of for example MlWin? I know you shouldn't use them because the distribution of the estimator isn't symmetric blablabla, but I need a measure of the variance of those estimates for pooling my multiple imputation results. Regards, Roel. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] anova on binomial LMER objects
The issues with lmer and the analysis of variance are due to its not make appropriate correction for the prior weights vector. If you convert your binomial response to the equivalent number of binary responses you get an appropriate anova table. It's on the ToDo list to fix this but a few other things have to come first, like grading assignments in one of my courses and repairing the computer in my office. This is the third motherboard I have torched in four months. On 9/29/05, Alan Cobo-Lewis [EMAIL PROTECTED] wrote: On Wed, 28 Sep 2005, Robert Bagchi wrote: Hi Patrick thanks for your advice. I have now tried glmmPQL, and it worked fine - I'm getting consistent results between plots and models fitted by glmmPQL. Plus it allows predict() and resid() which is another advantage over lmer at present. quick question though: why does one need to use PQL for binomial models? Is there a good reference for this? You don't have to use PQL for binomial models, but you can't use least-squares. PQL is an approximate solution. Laplace and Adaptive Gaussian Quadrature options in lmer are better approximations. So lmer would likely become the better option as it progresses in its development (though the current issues you've found with the F ratios certainly sound like maybe lmer isn't better for you in its current incarnation). alan -- Alan B. Cobo-Lewis, Ph.D. (207) 581-3840 tel Department of Psychology(207) 581-6128 fax University of Maine Orono, ME 04469-5742[EMAIL PROTECTED] http://www.umaine.edu/visualperception __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Saving Graphics
Hello all, I'm having difficulty automatically saving graphs. Is there a way to save graphs from the graphics window using commands in the R console? Thanks very much. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Problem with memory footprint of qq plot generated with lattice
Witold Eryk Wolski [EMAIL PROTECTED] wrote: Adding f.value=fn as argument to qqmath reduces the size of the image, but neither the axis (absicissae) nor the line added by panel.qqmathline are right. Adding f.value=fn as argument to panel.qqmathline and panel.qqmath generates the right graphic, but the size of the image is again 20 MB. Any Suggestions? Neither panel.qqmathline nor panel.qqmath pay any attention to an f.value argument. It looks like panel.qqmathline() effectively assumes that f.value is uniformly distributed. I guess you could rewrite panel.qqmathline and prepanel.qqmathline but I don't see a clean solution. -- Dave __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] solution of convolution equation
Hello, May be somebody can help me... I am trying to find a solution of a convolution equation using fft (and unfortunately I do not have a good background for this). So I am just trying to figure out how it can be implemented in R. I have two multidimensional independent variables X and Z and I know their densities fx and fz, which are multidimensional arrays. So I have to find the density of Y, such that X+Y=Z. So, first I tried on a simple example, where the variables are one-dimensional, say X is N(0,1) and Z is N(7,3). So I want to find the density of Y (which should be N(7, sqrt(8)). I did: x - seq(-6, 20, length=500) fx - dnorm(x) z - seq(-6, 20, length=500) fz - dnorm(z, mean=7, sd=3) ffty - fft(fz)/fft(fx) fy - fft(ffty, inverse=T)/length(ffty) plot(Re(fy)) But the plot is far from being normal. May be the problem is with the lengths of fx and fz? should they be of different lengths and fx padded with zeros? May be somebody could point out that…? Thanks! Anna __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Saving Graphics
See ?pdf or ?postscript -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mike Jones Sent: Thursday, September 29, 2005 3:22 PM To: r-help@stat.math.ethz.ch Subject: [R] Saving Graphics Hello all, I'm having difficulty automatically saving graphs. Is there a way to save graphs from the graphics window using commands in the R console? Thanks very much. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] lmer random effect model matrix question
I would create two 0/1 variables for sor level 1 and sor level 2 and use those as in mark$sor1 - ifelse(mark$sor == 1, 1, 0) mark$sor2 - ifelse(mark$sor == 2, 1, 0) (fm1 - lmer(y ~ sor + (0+sor1|lot) + (0+sor2|lot) + (1|wafer:lot), mark)) Linear mixed-effects model fit by REML Formula: y ~ sor + (0 + sor1 | lot) + (0 + sor2 | lot) + (1 | wafer:lot) Data: mark AIC BIClogLik MLdeviance REMLdeviance 455.7631 469.4231 -221.8816 453.5174 443.7631 Random effects: GroupsNameVariance Std.Dev. wafer:lot (Intercept) 35.866 5.9888 lot sor2222.709 14.9234 lot sor1 17.076 4.1323 Residual 12.569 3.5453 # of obs: 72, groups: wafer:lot, 24; lot, 8; lot, 8 Fixed effects: Estimate Std. Error DF t value Pr(|t|) (Intercept) 1995. 2.7581 70 723.3703 2e-16 *** sor2 10.0833 8.1622 70 1.2354 0.2208 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 On 9/29/05, Mark Lyman [EMAIL PROTECTED] wrote: I have one fixed effect, sor, with two levels. I have eight lots and three wafers from each lot. I have included the data below. I would like to fit a mixed model that estimates a covariance parameter for wafer, which is nested in lot, and two covariance parameters for lot, one for each level of sor. The following command fits the model that I want, except for it estimates the correlation between the two covariance parameters for lot. Is there anyway to make R not estimate this correlation? Thank you. lmer(y~sor+(sor-1|lot)+(1|wafer:lot),wafer) For those familiar with proc mixed the following SAS code fits the model that I want: proc mixed scoring=4; class sor lot wafer site; model y= sor/ddfm=satterth; random lot(sor)/group=sor; random wafer(lot); run; sor lot wafer sitey 11 1 11 2006 21 1 12 1999 31 1 13 2007 41 1 21 1980 51 1 22 1988 61 1 23 1982 71 1 31 2000 81 1 32 1998 91 1 33 2007 10 1 2 11 1991 11 1 2 12 1990 12 1 2 13 1988 13 1 2 21 1987 14 1 2 22 1989 15 1 2 23 1988 16 1 2 31 1985 17 1 2 32 1983 18 1 2 33 1989 19 1 3 11 2000 20 1 3 12 2004 21 1 3 13 2004 22 1 3 21 2001 23 1 3 22 1996 24 1 3 23 2004 25 1 3 31 1999 26 1 3 32 2000 27 1 3 33 2002 28 1 4 11 1997 29 1 4 12 1994 30 1 4 13 1996 31 1 4 21 1996 32 1 4 22 2000 33 1 4 23 2002 34 1 4 31 1987 35 1 4 32 1990 36 1 4 33 1995 37 2 5 11 2013 38 2 5 12 2004 39 2 5 13 2009 40 2 5 21 2023 41 2 5 22 2018 42 2 5 23 2010 43 2 5 31 2020 44 2 5 32 2023 45 2 5 33 2015 46 2 6 11 2032 47 2 6 12 2036 48 2 6 13 2030 49 2 6 21 2018 50 2 6 22 2022 51 2 6 23 2026 52 2 6 31 2009 53 2 6 32 2010 54 2 6 33 2011 55 2 7 11 1984 56 2 7 12 1993 57 2 7 13 1993 58 2 7 21 1992 59 2 7 22 1992 60 2 7 23 1990 61 2 7 31 1996 62 2 7 32 1993 63 2 7 33 1987 64 2 8 11 1996 65 2 8 12 1989 66 2 8 13 1996 67 2 8 21 1997 68 2 8 22 1993 69 2 8 23 1996 70 2 8 31 1990 71 2 8 32 1989 72 2 8 33 1992 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Saving Graphics
savePlot has been working quite well for me. Ken ~~~ Kenneth B. Pierce Jr. Research Ecologist Forestry Sciences Laboratory 3200 SW Jefferson Way Corvallis, OR 97331 [EMAIL PROTECTED] 541 750-7393 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mike Jones Sent: Thursday, September 29, 2005 12:22 PM To: r-help@stat.math.ethz.ch Subject: [R] Saving Graphics Hello all, I'm having difficulty automatically saving graphs. Is there a way to save graphs from the graphics window using commands in the R console? Thanks very much. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] multidimensional integration
Which version of R on which operating system crashes with which version of adapt? What are you trying to integrate? Is it pure R, or do you link to something else like C++? RSitSearch(integrate in 2 dimensions) produced 10 hits. Other searches produced more. My bottom line is that adapt has been around for a while. Can you provide a reproducible example of your problem -- preferably one as simple as possible, following the posting guide, www.R-project.org/posting-guide.html. (I believe that people who follow the posting guide usually get quicker, more useful answers.) spencer graves Marcel Prokopczuk wrote: dear all, i have the following problem: i want to integrate a two-dimensional function. unfortunately R crashes when i try to use adapt() and i get a nice windows message with some hex-code. do anybody of you knows how to avoid this or knows another more stable function than adapt() thanks a lot marcel __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] p-level in packages mgcv and gam
OK, I think I understand better but still have two points to clarify. The first one is about the number of df. I think those who replied on this objected to the way I chose df, not the fact that I would run a model with 7.4 df per se. If I read you correctly, I artificially reduce my p-value by using the estimated df found in a mgcv gam model into another model where I fix df. This is fine, I am quite willing not to run a second model with a fixed df and instead tell my readers that my model is marginally significant with a p-value of 0.03. This being said, do you know of guidelines for choosing df? A colleague told me he does not go above 10% of the number of points. Should I be concerned when mgcv estimates 7.4 df for 34 points? Note that for this particular model P 1e-16, and P is also very small if I fix df to either 4 or 7. My second point is the difference between models fitted by packages gam and mgcv. Sure, some of you have said different algorithms. And when I specify dfs, shouldn't P-values be very similar for the 2 packages? If not, what does it say of the confidence we can have in the models? I draw your attention to this exampl: I obtained P-values of 0.50 and 0.03 with packages gam and mgcv respectively: library(gam) Loading required package: splines data(kyphosis) kyp1 - gam(Kyphosis ~ s(Number, 3), family=binomial, data=kyphosis) summary.gam(kyp1) Call: gam(formula = Kyphosis ~ s(Number, 3), family = binomial, data = kyphosis) Deviance Residuals: Min 1Q Median 3Q Max -1.3646 -0.6233 -0.4853 -0.3133 2.0965 (Dispersion Parameter for binomial family taken to be 1) Null Deviance: 83.2345 on 80 degrees of freedom Residual Deviance: 71.9973 on 76. degrees of freedom AIC: 79.9976 Number of Local Scoring Iterations: 7 DF for Terms and Chi-squares for Nonparametric Effects Df Npar Df Npar Chisq P(Chi) (Intercept) 1 s(Number, 3) 1 21.37149 0.50375 detach(package:gam) library(mgcv) This is mgcv 1.3-7 kyp2 - gam(Kyphosis ~ s(Number, k=4, fx=T), family=binomial, data=kyphosis) summary.gam(kyp2) Family: binomial Link function: logit Formula: Kyphosis ~ s(Number, k = 4, fx = T) Parametric coefficients: Estimate Std. Error z value Pr(|z|) (Intercept) -1.5504 0.3342 -4.64 3.49e-06 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Approximate significance of smooth terms: edf Est.rank Chi.sq p-value s(Number) 33 8.898 0.0307 * --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 R-sq.(adj) = 0.101 Deviance explained = 12.5% UBRE score = 0.075202 Scale est. = 1 n = 81 kyp2$deviance [1] 72.85893 kyp2$null.deviance [1] 83.23447 kyp2$df.null [1] 80 kyp2$df.residual [1] 77 How can we explain this huge difference? Denis Le 05-09-29 à 06:00, [EMAIL PROTECTED] a écrit : De : Henric Nilsson [EMAIL PROTECTED] Date : 29 septembre 2005 03:55:19 HAE À : [EMAIL PROTECTED] Cc : r-help@stat.math.ethz.ch Objet : Rép : [R] p-level in packages mgcv and gam Répondre à : Henric Nilsson [EMAIL PROTECTED] Yves Magliulo said the following on 2005-09-28 17:05: hi, i'll try to help you, i send a mail about this subject last week... and i did not have any response... I'm using gam from package mgcv. 1) How to interpret the significance of smooth terms is hard for me to understand perfectly : using UBRE, you fix df. p-value are estimated by chi-sq distribution using GCV, the best df are estimated by GAM. (that's what i want) and p-values This is not correct. The df are estimated in both cases (i.e. UBRE and GCV), but the scale parameter is fixed in the UBRE case. Hence, by default UBRE is used for family = binomial or poisson since the scale parameter is assumed to be 1. Similarly, GCV is the default for family = gaussian since we most often want the scale (usually denoted sigma^2) to be estimated. are estimated by an F distribution But in that case they said use at your own risk in ?summary.gam The warning applies in both cases. The p-values are conditional on the smoothing parameters, and the uncertainty of the smooths is not taken into account when computing the p-values. so you can also look at the chi.sq : but i don't know how to choose a No... criterion like for p-values... for me, chi.sq show the best predictor in a model, but it's hard to reject one with it. Which version of mgcv do you use? The confusion probably stems from earlier versions of mgcv ( 1.3-5): the summary and anova methods used to have a column denoted Chi.sq even when the displayed statistic was computed as F. Recent versions of mgcv has summary(b) Family: gaussian Link function: identity Formula: y ~ s(x0) + s(x1) + s(x2) + s(x3) Parametric coefficients: Estimate Std. Error t value Pr(|t|) (Intercept) 7.9150 0.1049
Re: [R] solution of convolution equation
You need to pad both fx and fz with zeros to at least a length of length(fx) + length(fz) - 1, because you're really computing a circular convolution. The same holds in higher dimensions (for each dimension). Reid Huntsinger -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Anna Oganyan Sent: Thursday, September 29, 2005 3:23 PM To: r-help@stat.math.ethz.ch Subject: [R] solution of convolution equation Hello, May be somebody can help me... I am trying to find a solution of a convolution equation using fft (and unfortunately I do not have a good background for this). So I am just trying to figure out how it can be implemented in R. I have two multidimensional independent variables X and Z and I know their densities fx and fz, which are multidimensional arrays. So I have to find the density of Y, such that X+Y=Z. So, first I tried on a simple example, where the variables are one-dimensional, say X is N(0,1) and Z is N(7,3). So I want to find the density of Y (which should be N(7, sqrt(8)). I did: x - seq(-6, 20, length=500) fx - dnorm(x) z - seq(-6, 20, length=500) fz - dnorm(z, mean=7, sd=3) ffty - fft(fz)/fft(fx) fy - fft(ffty, inverse=T)/length(ffty) plot(Re(fy)) But the plot is far from being normal. May be the problem is with the lengths of fx and fz? should they be of different lengths and fx padded with zeros? May be somebody could point out that...? Thanks! Anna __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Fisher's discriminant functions
This are in various packages, you could have a look at ade4 (on CRAN). Kjetil C NL wrote: Hi everyone, I'm trying to solve a problem about how to get the Fisher's discriminant functions of a lda (linear discriminant analysis) object, I mean, the object obtained from doing lda(formula, data) function of the package MASS in R-project. This object gives me the canonical linear functions (n-1 coefficients matrix of n groups at least), and only with this information I could predict the group of an observation data using the predict function. But what I need is the Fisher's discriminant functions (n coefficients matrix of n groups) in order to classify my future data. The object predict gives me only the following attributes x, posterior and class, but none of them are the coefficients matrix of the Fisher's discriminant functions, and the reason why I'm not using the predict function for my predictions is because the time spent is very high for what I'm expecting, about 0.5 seconds while I can obtain this prediction with the Fisher's discriminant functions faster. So, I don't know if there's a package which I can use to obtain the mentioned coefficients matrix of the Fisher's discriminant functions. I anyone can help, I would appreciate it greatly. Thank you and regards. Carlos Niharra López __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] solution of convolution equation
Anna Oganyan wrote: Hello, May be somebody can help me... I am trying to find a solution of a convolution equation using fft (and unfortunately I do not have a good background for this). So I am just trying to figure out how it can be implemented in R. I have two multidimensional independent variables X and Z and I know their densities fx and fz, which are multidimensional arrays. So I have to find the density of Y, such that X+Y=Z. So, first I tried on a simple example, where the variables are one-dimensional, say X is N(0,1) and Z is N(7,3). So I want to find the density of Y (which should be N(7, sqrt(8)). I did: x - seq(-6, 20, length=500) fx - dnorm(x) z - seq(-6, 20, length=500) fz - dnorm(z, mean=7, sd=3) ffty - fft(fz)/fft(fx) fy - fft(ffty, inverse=T)/length(ffty) plot(Re(fy)) But the plot is far from being normal. May be the problem is with the lengths of fx and fz? should they be of different lengths and fx padded with zeros? May be somebody could point out that…? Thanks! Anna __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Hello Anna, in our R package distr (on CRAN) we have implemented a convolution algorithm via fft. See also: http://www.uni-bayreuth.de/departments/math/org/mathe7/KOHL/pubs/comp.pdf respectively library(distr) getMethod(+, signature=c(AbscontDistribution,AbscontDistribution)) (or see the sources) Unfortunatelly, we haven't implemented our algorithms for multidimensional distributions so far. hope that helps, Matthias __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] xyplots
On 9/28/05, Nathan Leon Pace, MD, MStat [EMAIL PROTECTED] wrote: A related question: My xyplot is essentially a time series (up-and-down experimental design). Thus I need to connect the points sequentially regardless of the group value. With the groups argument, type = 'b' gives two lines - one for each group. xyplot(rnorm(100) ~ 1:100, groups = sample(1:2, 100, TRUE), panel = function(x, y, ..., type) { panel.lines(x, y, ...) panel.superpose(x, y, ..., type = 'p') }) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Bug in lmer?
I am relatively new to R so I am not confident enough in what I am doing to be certain this is a bug. I am running R 2.1.1 on a Windows XP machine and the lme4 package version 0.98-1. The following code fits the model I want using the nlme package version 3.1-60. mltloc$loc - factor(mltloc$loc) mltloc$block - factor(mltloc$block) mltloc$trt - factor(mltloc$trt) Mltloc - groupedData(adg~trt|loc,mltloc) plot(Mltloc) plot(Mltloc,outer=~trt) lme(adg~trt, random=pdBlocked(list(pdIdent(~1),pdIdent(~block-1),pdIdent(~trt-1))),Mltloc) The problem is that when I try fitting the model using the lmer function with the following code: lmer(adg~trt+(1|loc)+(1|block:loc)+(1|loc:trt),mltloc) I get this message from Windows and R closes. R for Windows GUI front-end has encountered a problem and needs to close. We are sorry for the inconvenience. This same code works on a Macintosh. So it doesn't seem that I have made an error in my code. Also if anyone of the random effect terms is removed there is no problem. Is this something that is being looked at? Or I have I made a mistake somewhere? I have included the data that I am using below. X obs loc block trt adg fe 1 1 3 A 1 3 3.16454 7.1041 2 2 4 A 1 4 3.12500 6.6847 3 3 6 A 1 2 3.15944 6.8338 4 4 7 A 1 1 3.25000 6.5254 5 5 9 A 2 2 2.71301 8.2505 6 6 10 A 2 1 3.20281 7.5922 7 7 12 A 2 3 3.02423 7.3894 8 8 16 A 2 4 2.87245 7.4604 9 9 19 A 3 1 2.68878 8.2785 10 10 20 A 3 2 2.86862 7.9470 11 11 21 A 3 3 2.89923 7.9739 12 12 22 A 3 4 3.02806 8.4331 13 13 25 B 1 3 2.18131 6.6691 14 14 27 B 1 4 2.51914 5.6281 15 15 28 B 1 2 1.88739 7.0723 16 16 31 B 1 1 2.34685 6.0295 17 17 33 B 2 4 2.45608 5.6195 18 18 35 B 2 1 2.25225 6.3978 19 19 36 B 2 3 2.23649 6.1799 20 20 40 B 2 2 2.47523 5.9985 21 21 41 B 3 2 1.94200 7.2975 22 22 44 B 3 3 2.43243 6.4350 23 23 47 B 3 4 2.30180 6.3339 24 24 48 B 3 1 2.53378 6.1564 25 25 50 C 1 4 2.96014 7.5110 26 26 51 C 1 2 3.23551 7.4762 27 27 54 C 1 3 3.24638 7.2063 28 28 56 C 1 1 3.04710 7.6389 29 29 58 C 2 3 3.26449 7.5466 30 30 59 C 2 2 2.71377 9.0895 31 31 61 C 2 1 3.06522 7.8723 32 32 62 C 2 4 2.71739 8.2318 33 33 66 C 3 4 3.03623 7.9426 34 34 67 C 3 3 3.10507 8.4608 35 35 69 C 3 1 3.16304 8.5549 36 36 70 C 3 2 3.02899 8.5038 37 37 74 D 1 1 2.49164 9.5758 38 38 77 D 1 3 2.51833 9.5121 39 39 79 D 1 2 2.35631 10.3264 40 40 80 D 1 4 2.30331 9.7715 41 41 81 D 2 3 2.72688 9.5628 42 42 83 D 2 2 2.59512 9.9414 43 43 85 D 2 1 2.56516 9.3887 44 44 88 D 2 4 2.91523 8.3158 45 45 89 D 3 3 2.57943 10.4416 46 46 90 D 3 4 2.98159 8.7710 47 47 93 D 3 2 2.35370 11.0148 48 48 94 D 3 1 2.21953 11.2417 49 49 99 E 1 3 2.84158 8.7886 50 50 100 E 1 4 2.65264 8.6946 51 51 102 E 1 2 2.47112 9.7143 52 52 103 E 1 1 2.89769 9.2401 53 53 105 E 2 2 2.57343 9.5353 54 54 106 E 2 1 2.99752 8.7538 55 55 110 E 2 4 2.95380 8.8210 56 56 112 E 2 3 3.08663 8.9427 57 57 114 E 3 1 2.72525 9.4308 58 58 115 E 3 2 2.75825 9.7721 59 59 116 E 3 3 3.08333 8.9010 60 60 117 E 3 4 3.12129 8.4852 61 61 122 F 1 1 3.20600 6.3983 62 62 123 F 1 2 2.89500 6.6569 63 63 125 F 1 4 3.36900 6.0821 64 64 126 F 1 3 3.12000 6.5349 65 65 130 F 2 2 3.19300 6.6729 66 66 131 F 2 1 3.29800 6.5488 67 67 133 F 2 4 3.09700 6.6598 68 68 135 F 2 3 3.38500 6.2998 69 69 139 F 3 3 3.44900 6.2849 70 70 140 F 3 2 3.05000 6.9957 71 71 141 F 3 1 3.43500 6.7302 72 72 143 F 3 4 3.60600 6.3827 73 73 145 G 1 2 2.58669 8.1394 74 74 147 G 1 1 3.17892 7.0972 75 75 148 G 1 4 2.95284 7.3140 76 76 151 G 1 3 3.17924 6.9430 77 77 154 G 2 4 2.62344 7.5150 78 78 155 G 2 3 2.64286 8.0237 79 79 157 G 2 1 3.12760 7.3169 80 80 160 G 2 2 2.54993 8.1957 81 81 163 G 3 4 2.58322 7.9687 82 82 164 G 3 3 2.84813 7.9284 83 83 166 G 3 2 2.69279 8.5303 84 84 167 G 3 1 3.14424 7.3564 85 85 169 H 1 3 3.39974 6.5945 86 86 173 H 1 2 3.12370 6.7530 87 87 175 H 1 4 3.17969 6.4279 88 88 176 H 1 1 3.70052 6.4830 89 89 177 H 2 2 2.95192 7.3809 90 90 179 H 2 3 3.44661 6.7929 91 91 182 H
Re: [R] standard error of variances and covariances of the random effects with LME
Doran, Harold wrote: You cannot. Yes. But when it is really needed, as the original poster said, what would be wrong with taking the length of a 95% confidence interval and dividing into 4? (of course it will be wrong, but so much as to be useless?) Kjetil Also, it's not that the distribution of the random effects is not symmetric, but that it *may* not be symmetric, and this is an assumption that should be checked. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roel de Jong Sent: Thursday, September 29, 2005 9:20 AM To: r-help@stat.math.ethz.ch Subject: [R] standard error of variances and covariances of the random effects with LME Hello, how do I obtain standard errors of variances and covariances of the random effects with LME comparable to those of for example MlWin? I know you shouldn't use them because the distribution of the estimator isn't symmetric blablabla, but I need a measure of the variance of those estimates for pooling my multiple imputation results. Regards, Roel. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Select varying LS digits in long numbers?
I've received what looks like a good -- and neat! -- solution off-list from Hadley Wickham. Suppose numbers is a vector of long integers. Then Hadley wrote: digits - outer(numbers, 10:0, function(x,y) numbers %/% 10^y %% 10) apply(digits,2, function(x) length(unique(x))) This certainly looks cast-iron, and nicely passes my slipperiness test: numbers-c(1234566999,1234567001) digits - outer(numbers, 10:0, function(x,y) numbers %/% 10^y %% 10) result-apply(digits,2, function(x) length(unique(x))); result # [1] 1 1 1 1 1 1 1 2 2 2 2 d1-min(which(result1)); d1 # [1] 8 numbers%%(10^(12-d1)) # [1] 6999 7001 as desired! Thanks also to Patrick Burns and Jim Holtman for other suggestions based (in effect) on diff(range(numbers)). On 29-Sep-05 Ted Harding wrote: Hi Folks, I'm trying to find a neat solution to an apparently simple problem, but one which turns out to be a bit more intricate and tricky than one might expect. Suppose I have numbers given to a large number of digits. For example 1234567021 where (though I don't know this beforehand) only the last 3 digits will be varying (and all 3 will vary). What I want is, give a vector x of such numbers, to extract the minimal set of final digits which will include the varying digits (i.e. in this case the last 3 digits). And there may be a decimal point somewhere along the line (though again I won't know where, nor whether). I can think of brute-force ways of doing this, but I'd like a neat one! Best wishes to all, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 29-Sep-05 Time: 18:45:26 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 29-Sep-05 Time: 22:35:15 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Binary Logit Regression with R
Hi to all, I am a PH.D Student doing statistical analysis. I am totally new to R. I previously use Stata and am changing into R. I ususally do with logit regression with binary dependent variable (war occurence:1 or 0). I just want to know command to do that. More sepcifically, Let say, my Y is war occurence (occur=1, otherwise 0). And my independent variables (Xs) are trade, democracy, military poweretc. In Stata, I do like what follows: logit war trade democracy militarypower... Then I will get results. What are the equivalent command in R? Many thanks, JP __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Binary Logit Regression with R
On Thu, 2005-09-29 at 18:08 -0400, Johann Park wrote: Hi to all, I am a PH.D Student doing statistical analysis. I am totally new to R. I previously use Stata and am changing into R. I ususally do with logit regression with binary dependent variable (war occurence:1 or 0). I just want to know command to do that. More sepcifically, Let say, my Y is war occurence (occur=1, otherwise 0). And my independent variables (Xs) are trade, democracy, military poweretc. In Stata, I do like what follows: logit war trade democracy militarypower... Then I will get results. What are the equivalent command in R? Many thanks, JP See ?glm in the base stats package or ?lrm in Frank Harrell's Design package on CRAN. BTW, doing: help.search(logit) or RSiteSearch(logit) would provide you with the ability to do keyword searches of your current R installation or the online e-mail list and documentation archives, respectively. You should also review Chapter 11 - Statistical Models in R in An Introduction to R, which is installed with R or available online under the Documentation/Manuals link on the main R web site. HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Binary Logit Regression with R
Hi, Johann Park wrote: Let say, my Y is war occurence (occur=1, otherwise 0). And my independent variables (Xs) are trade, democracy, military poweretc. Take a look at ?glm. HTH, Kev -- Ko-Kang Kevin Wang PhD Student Centre for Bioinformation Science Building 27, Room 1004 Mathematical Sciences Institute (MSI) Australian National University Canberra, ACT 2601 Australia Homepage: http://wwwmaths.anu.edu.au/~wangk/ Ph (W): +61-2-6125-2431 Ph (H): +61-2-6125-7488 Ph (M): +61-40-451-8301 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Display values in piechart/barplot
Volker Rehbock wrote: Is it possible to automatically display the underlying values of a piechart/barplot in the graphic? If so, which package/function/argument do I need for it? floating.pie and pie.labels in the plotrix package. Jim __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] reshaping data?
Karin Lagesen wrote: I have a file like this: a 0.1 a 0.2 a 0.9 b 0.5 b 0.9 b 0.7 c 0.6 c 0.99 c 0.88 Which I would like to get to be the following matrix: 0.1 0.20.30.4 ... a 12 0 0 b 00 0 0 .. If you have a file named KL.dat with the above data in it: KL.df-read.table(KL.dat,header=FALSE) KL.mat-as.matrix(table(KL.df)) Jim __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] standard error of variances and covariances of the random effects with LME
Well I finally figured it out. If you use the delta method with transformations exp(x) for the standard deviations and ((exp(y)-1)/(exp(y)+1) for the correlation elements of the apVar structure, which is btw *not* the inverse of a fisher transformation, you get the standard errors. Douglas Bates wrote: With lme you could but it would take a while to work it out. There is an approximate Hessian matrix for the parameters that determine the variance-covariance matrix in there somewhere and if you were sufficiently persistent you could convert that apVar component to the scale of the variances and covariances. I believe it is in the scale of the logarithm of the standard deviation and Fisher's z transformation (i.e. the hyperbolic arc tangent) of the correlation. On 9/29/05, Doran, Harold [EMAIL PROTECTED] wrote: You cannot. Also, it's not that the distribution of the random effects is not symmetric, but that it *may* not be symmetric, and this is an assumption that should be checked. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roel de Jong Sent: Thursday, September 29, 2005 9:20 AM To: r-help@stat.math.ethz.ch Subject: [R] standard error of variances and covariances of the random effects with LME Hello, how do I obtain standard errors of variances and covariances of the random effects with LME comparable to those of for example MlWin? I know you shouldn't use them because the distribution of the estimator isn't symmetric blablabla, but I need a measure of the variance of those estimates for pooling my multiple imputation results. Regards, Roel. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] mvtnorm package
Hi all, I've been trying to install the mvtnorm package (in a Linux R version) without sucess. I write install.packages(mvtnorm,lib=/home/posmae/cnaber,repos=http://cran.uk.r-project.org/;) and the following message arises == downloaded 160Kb * Installing *source* package 'mvtnorm' ... ** libs g77 -fPIC -g -O2 -c mvt.f -o mvt.o gcc -I/usr/lib/R/include -fPIC -g -O2 -c randomF77.c -o randomF77.o gcc -shared -o mvtnorm.so mvt.o randomF77.o -lg2c -lm -lgcc_s -L/usr/lib/R/lib -lR /usr/bin/ld: cannot find -lg2c collect2: ld returned 1 exit status make: ** [mvtnorm.so] Erro 1 ERROR: compilation failed for package 'mvtnorm' ** Removing '/home/posmae/cnaber/mvtnorm' ** Restoring previous '/home/posmae/cnaber/mvtnorm' The downloaded packages are in /tmp/Rtmp25z35w/downloaded_packages warning message: installation of package 'mvtnorm' had non-zero exit status in: install.packages(mvtnorm, lib = /home/posmae/cnaber, repos = http://cran.uk.r-project.org/;) I don't know what's the problem because I've installed the package sn with sucess Thanks all, Caio - [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Code to read a SAS file
Hi, It is possible you provide me the code to read a SAS file in R. I tried the read.ssd but I didn't get very well on it. Many thanks. M. - Comprueba qué es nuevo, aquí [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] mvtnorm package
On 30 September 2005 at 01:55, Caio Lucidius Naberezny Azevedo wrote: | == | downloaded 160Kb | * Installing *source* package 'mvtnorm' ... | ** libs | g77 -fPIC -g -O2 -c mvt.f -o mvt.o | gcc -I/usr/lib/R/include -fPIC -g -O2 -c randomF77.c -o randomF77.o | gcc -shared -o mvtnorm.so mvt.o randomF77.o -lg2c -lm -lgcc_s -L/usr/lib/R/lib -lR | /usr/bin/ld: cannot find -lg2c That's the error. If you're on Debian, you need libg2c0-dev. But on Debian you may as just install the existing binary package r-cran-mvtnorm. If you're not on Debian you need to figure out where to get the complete set of your libg2c tools from. | I don't know what's the problem because I've installed the package sn with sucess Sn may not contain fortran sources. Dirk -- Statistics: The (futile) attempt to offer certainty about uncertainty. -- Roger Koenker, 'Dictionary of Received Ideas of Statistics' __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html