Re: [R] parallel computing package on a multicore windows workstation
On Mon, 6 Jun 2011, Richard Wang wrote: Hi, I would like to get suggestion about parallel computing package on a multicore windows workstation. I tried doSMP, but it crashes R a lot. I am wondering if snow and snowfall can be used on a single workstation (i,e, not cluster). At suggestion would be appreciated, Yes. See for example packages RSiena and boot (version 1.3-1). (I presume you meant to say 'Windows', the OS: it is capitalized. On other OSes I would suggest package multicore as easier to use than snow.) Best, Richard [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, rip...@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Can't compile rtiff for OSX
On Sun, 5 Jun 2011, Robert Baruch wrote: I have libtiff installed from macports, and it is sitting there in /opt/local/lib: Well, you haven't told us which build of R you used, but it looks like it is not the one from MacPorts (maybe the one from CRAN?). /opt/local is not a standard location on Mac OS X (nor any other Unix-alike that I have access to). Please see the posting guide: this is a not a suitable question for R-help, and you are asked to post question about Mac OS X on r-sig-mac. The answer involves 'discussion unintelligible to non-programmers' to quote the posting guide. The version of tiff which matches the CRAN build of R is at http://r.research.att.com/libs/ -- it works for me. $ ls -l /opt/local/lib/*tiff* -rwxr-xr-x 2 root admin 796684 Jun 4 22:13 /opt/local/lib/libtiff.3.dylib -rw-r--r-- 2 root admin 990296 Jun 4 22:13 /opt/local/lib/libtiff.a lrwxr-xr-x 1 root admin 15 Jun 4 22:14 /opt/local/lib/libtiff.dylib - libtiff.3.dylib -rwxr-xr-x 2 root admin 960 Jun 4 22:13 /opt/local/lib/libtiff.la -rwxr-xr-x 2 root admin 35400 Jun 4 22:13 /opt/local/lib/libtiffxx.3.dylib -rw-r--r-- 2 root admin 17280 Jun 4 22:13 /opt/local/lib/libtiffxx.a lrwxr-xr-x 1 root admin 17 Jun 4 22:14 /opt/local/lib/libtiffxx.dylib - libtiffxx.3.dylib -rwxr-xr-x 2 root admin 998 Jun 4 22:13 /opt/local/lib/libtiffxx.la $ ls -l /opt/local/include/*tiff* -rw-r--r-- 2 root admin 33813 Jun 4 22:13 /opt/local/include/tiff.h -rw-r--r-- 2 root admin 3020 Jun 4 22:13 /opt/local/include/tiffconf.h -rw-r--r-- 2 root admin 20514 Jun 4 22:13 /opt/local/include/tiffio.h -rw-r--r-- 2 root admin 1706 Jun 4 22:13 /opt/local/include/tiffio.hxx -rw-r--r-- 2 root admin410 Jun 4 22:13 /opt/local/include/tiffvers.h However, when I try to install rtiff, R is not picking up libtiff: $ R CMD INSTALL ~/Downloads/rtiff_1.4.1.tar.gz * installing to library ‘/Library/Frameworks/R.framework/Versions/2.13/Resources/library’ * installing *source* package ‘rtiff’ ... checking for gcc... gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ANSI C... none needed checking for TIFFOpen in -ltiff... no Error: Could not find libtiff. ERROR: configuration failed for package ‘rtiff’ * removing ‘/Library/Frameworks/R.framework/Versions/2.13/Resources/library/rtiff’ TIFFOpen is in libtiff: $ nm /opt/local/lib/libtiff.dylib | grep TIFFOpen 00030740 T _TIFFOpen So I think the conclusion is that R isn't even looking in /opt/local/lib. Seems like a basic question, but how do I get R to look in /opt/local/{lib|include}? Thanks for any help! --Rob __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, rip...@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Question about curve function
On Sun, 5 Jun 2011, Abhilash Balakrishnan wrote: Dear Mr. Murdoch, I find out that still do not understand why the following does not work: curve(expression(x)) Error in xy.coords(x, y, xlabel, ylabel, log) : 'x' and 'y' lengths differ As here the input to curve is an expression, as documented in the help, and Not really, and certainly not in the sense you seem to understand it.. 'expression(x)' is a call to the expression() function, and that evaluates to a length-one expression vector. As ?expression says: ‘Expression’ here is not being used in its colloquial sense, that of mathematical expressions. Those are calls (see ‘call’) in R, and an R expression vector is a list of calls, symbols etc, for example as returned by ‘parse’. the expression is simply x. 'Simply' untrue. What is the y mentioned in the error? There is no y used here. Yes, there is. Please do read the code for 'curve': y - eval(expr, envir = list(x = x), enclos = parent.frame()) so you are trying to plot a length-1 expression vector against a length-101 'x'. As others have said, curve() is a convenience function, and its requirements are rather picky. And you have already been given one good solution, curve(I). Thank you for support. Abhilash B. On Sun, Jun 5, 2011 at 3:39 PM, Duncan Murdoch murdoch.dun...@gmail.comwrote: On 11-06-05 1:07 PM, Abhilash Balakrishnan wrote: Dear Sirs, I am a new user of the R package. When I try to use the curve function it confuses me. curve(x^2) Works fine. curve(x) Makes a complaint I don't understand. Why is x^2 valid and x is not? curve() is a convenience function, and it tries to guess what you mean. Sometimes it gets it wrong. In the first case, it is clear you want to graph x^2. In the second it guesses you have a function named x and want to graph that. You don't, so it fails. Probably it could try again after the first failure, but I'd guess there will always be strange cases where it does weird things. Duncan Murdoch I check the documentation of curve, and it says the first argument must be an expression containing x. expression(x) Is an expression containing x. curve(expression(x)) Makes a different complaint and mentions different lengths of x and y (but I use no y here). I understand that plotting the function y(x) = x is rather silly, but I want to know what I am doing wrong, for the sake of my understanding of how R works. Thank you for support. Abhilash B. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, rip...@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] logistic growth model
Write the growth formula in an R script. Define initial par values. Input the size and age data. Plot the size and age data as points. Plot the growth model with the initial par values as a line. Play with the initial par values until you see a good agreement between the model (the line) and the data (the points). Optimise. Re-plot. Plot a residual histogram. Plot a residual scatterplot. Plot a Q-Q residual plot. HTH Rubén Dr. Rubén Roa-Ureta AZTI - Tecnalia / Marine Research Unit Txatxarramendi Ugartea z/g 48395 Sukarrieta (Bizkaia) SPAIN -Mensaje original- De: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] En nombre de Renalda Enviado el: sábado, 04 de junio de 2011 6:17 Para: r-help@r-project.org Asunto: [R] logistic growth model I want to Fit a logistic growth model for y = k *eb0+b1(age)/1 + eb0+b1(age), can some one help on how to get the initial coefficients b0 and b1? I need to estimate in order to do the regression analysis. When I run using b0=0.5 and b1=3.4818, I get the following error 397443.8 : 0.5 3.4818 Error in nls(Height ~ k * exp(b1 + b2 * Age)/(1 + exp(b1 + b2 * Age)), : singular gradient please tell me what is wrong with my initials values, and how to get the initial values __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Can R do zero inflated gamma regression?
siriustar qinlangjinan at live.cn writes: Hi, Dear R-help I know there are some R package to deal with zero-inflated count data. But I am now looking for R package to deal with zero-inflated continuous data. The response variable (Y) in my dataset contains a larger mount of zero and the Non-zero response are quite right skewed. Now what i am doing is first to use a logistic regression on covariates (X) to estimate the probability of Y being 0. Then focus on the dataset where Y is not zero, and run a linear regression or gamma glm to estimate the association between Y and X when Y is not zero. However, the linear regression and gamma glm model fit my data poorly. So, I am thinking maybe a zero inflated gamma or zero inflated lognormal regression are helpful, where I can estimate the probability of Y being zero and the association between non zero Y and X at the same time. However, I dont know which R package can do that. I think your 'conditional' strategy is quite useful in general, and may in general give you the same answers as the zero-inflated approach you're suggesting. Perhaps there are some other issues with the conditional (gamma GLM) parts of your analysis? Have you tried simple log-linear regression (i.e. assuming that the non-zero values are lognormally distributed)? I would recommend reading this thread in the r-sig-ecology mailing list: http://thread.gmane.org/gmane.comp.lang.r.ecology/2124 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Question about curve function
As a further example of the trickiness, the function method of plot() relies on curve(x, ...) being a request to plot the function x(x) against x. I've added a comment to that effect to the help page. On Mon, 6 Jun 2011, Prof Brian Ripley wrote: On Sun, 5 Jun 2011, Abhilash Balakrishnan wrote: Dear Mr. Murdoch, I find out that still do not understand why the following does not work: curve(expression(x)) Error in xy.coords(x, y, xlabel, ylabel, log) : 'x' and 'y' lengths differ As here the input to curve is an expression, as documented in the help, and Not really, and certainly not in the sense you seem to understand it.. 'expression(x)' is a call to the expression() function, and that evaluates to a length-one expression vector. As ?expression says: ‘Expression’ here is not being used in its colloquial sense, that of mathematical expressions. Those are calls (see ‘call’) in R, and an R expression vector is a list of calls, symbols etc, for example as returned by ‘parse’. the expression is simply x. 'Simply' untrue. What is the y mentioned in the error? There is no y used here. Yes, there is. Please do read the code for 'curve': y - eval(expr, envir = list(x = x), enclos = parent.frame()) so you are trying to plot a length-1 expression vector against a length-101 'x'. As others have said, curve() is a convenience function, and its requirements are rather picky. And you have already been given one good solution, curve(I). Thank you for support. Abhilash B. On Sun, Jun 5, 2011 at 3:39 PM, Duncan Murdoch murdoch.dun...@gmail.comwrote: On 11-06-05 1:07 PM, Abhilash Balakrishnan wrote: Dear Sirs, I am a new user of the R package. When I try to use the curve function it confuses me. curve(x^2) Works fine. curve(x) Makes a complaint I don't understand. Why is x^2 valid and x is not? curve() is a convenience function, and it tries to guess what you mean. Sometimes it gets it wrong. In the first case, it is clear you want to graph x^2. In the second it guesses you have a function named x and want to graph that. You don't, so it fails. Probably it could try again after the first failure, but I'd guess there will always be strange cases where it does weird things. Duncan Murdoch I check the documentation of curve, and it says the first argument must be an expression containing x. expression(x) Is an expression containing x. curve(expression(x)) Makes a different complaint and mentions different lengths of x and y (but I use no y here). I understand that plotting the function y(x) = x is rather silly, but I want to know what I am doing wrong, for the sake of my understanding of how R works. Thank you for support. Abhilash B. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, rip...@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 -- Brian D. Ripley, rip...@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Error with BRugs 0.53 and 0.71, on Win7 with R 2.12.2 and 2.13.0 (crashes R GUI)
On 05.06.2011 23:42, Chris Chapman wrote: Thanks again. FWIW, I tried an even older R version (2.11) + BRugs 0.53 and 0.70 ... but got the same errors as with the iterations reported below. So I'm giving up on trying to solve the issue. My workaround is that I'm using the R2WinBUGS package instead. So far that has worked well -- I've run a few iterations of models with no problems. Specifically, none of the errors below have occurred (this is using R 2.13.0, R2WinBUGS 2.1-18, OpenBUGS 3.0.3 as installed by BRugs, running under Windows 7/32-bit). Unlike with BRugs, I'm able to use R2WinBUGS to run a model, change it, run again with new parameters, etc., with no crashes or other problems yet. Strange, since the BRugs example always worked well for me and some others. Hence I cannot tell where the interaction comes from that causes the crashes. Uwe: thank you for your work there as well! Well, I am very far behind my schedule for BRugs!!! Best wishes, uwe -- Chris -- From: Uwe Ligges lig...@statistik.tu-dortmund.de Sent: Monday, May 30, 2011 12:57 AM To: Chris Chapman cnchap...@msn.com Cc: r-help@r-project.org Subject: Re: [R] Error with BRugs 0.53 and 0.71, on Win7 with R 2.12.2 and 2.13.0 (crashes R GUI) On 29.05.2011 23:19, Chris Chapman wrote: Uwe -- thank you. No, this occurs on three different machines: two at work (a Lenovo laptop running Win7-32, plus an HP workstation running Win7-64) ... and I just tried another Compaq desktop machine at home running WinXP-32, with the same result. I agree that this seems highly unusual since the examples are so simple and obviously work in general; yet it is also highly replicable for me, and I'm at a loss as to what might be the root cause given the different machines and Windows versions. FWIW, everything else in my R environment (Rgui, Tinn-R, RStudio, ggplot2, various other packages) runs OK. OpenBUGS in itself also seems OK albeit in limited tests. Could there be something in the handleRes() error that suggests anything to examine (firewall, antivirus, file locations, or some file permissions, perhaps? -- although those also vary across my machines, esp. from work to home). The only thing I can say is, yes, perhaps. Sorry, but I am still unable to reproduce so far. Uwe Thanks again, -- Chris -- From: Uwe Ligges lig...@statistik.tu-dortmund.de Sent: Sunday, May 29, 2011 10:07 AM To: Chris Chapman cnchap...@msn.com Cc: r-help@r-project.org Subject: Re: [R] Error with BRugs 0.53 and 0.71, on Win7 with R 2.12.2 and 2.13.0 (crashes R GUI) Sounds like a hardware problem to me, since I do not expereice any problems with the example you gave at first. Is this all on the same hardware? Uwe Ligges On 27.05.2011 18:38, Chris Chapman wrote: I've run into persistent problems with OpenBUGS crashing when using BRugs .53 and .71, and am hoping someone has suggestions. There is obviously something unusual going on in my environment, but I'm at a loss as to where to begin to try to solve it. In a nutshell, what happens is that, as soon as I call modelCheck() in BRugs, it gets an error or crashes ... but only some of the time (90% p 100%). Following are details: 1. OpenBUGS 3.0.3 + BRugs 0.531: It works occasionally, but approximately 90% of the time, I get the following error from modelCheck(): Error in handleRes(res[[3]]) : An OpenBUGS module or procedure was called that did not exist. The specific code seems not to matter, but here is an example (model taken from the OpenBUGS tutorial): modelString = model { for (i in 1:N) { r[i] ~ dbin(p[i], n[i]) b[i] ~ dnorm(0, tau) logit(p[i])- alpha0 + alpha1 * x1[i] + alpha2 * x2[i] + alpha12 * x1[i] * x2[i] + b[i] } alpha0 ~ dnorm(0, 1.0E-6) alpha1 ~ dnorm(0, 1.0E-6) alpha2 ~ dnorm(0, 1.0E-6) alpha12 ~ dnorm(0, 1.0E-6) tau ~ dgamma(0.001, 0.001) sigma- 1 / sqrt(tau) } print(modelString) writeLines(modelString,con=model3.txt) modelCheck( model3.txt ) Which (usually) produces: modelCheck( model3.txt ) Error in handleRes(res[[3]]) : An OpenBUGS module or procedure was called that did not exist. I've copied at the end of this message an example from a single R session that shows how it may work sometimes but not others. 2. OpenBUGS 3.2.1 + BRugs 0.71: As above, the model occasionally works, but mostly it crashes R on the modelCheck() line with the error R for Windows GUI front-end has stopped working. I've tried the following combinations to try to get it to work: A. Win7 32-bit + R 2.13 + BRugs 0.531 from standard CRAN repository (installed from R) B. Win7 32-bit + R 2.13 + BRugs 0.71 + OpenBUGS 3.2.1 (package and EXE from OpenBUGS site) C. Win7 64-bit [different machine] + R 2.13 (32-bit) + BRugs 0.531 D. Win7 64-bit + R 2.13 (32-bit) + BRugs 0.71 + OpenBugs 3.2.1 E. Win7 32-bit + *R 2.12.2* + BRugs 0.531 from standard CRAN repository (installed from R) F. Win7 64-bit + R 2.12.2
Re: [R] logistic growth model
If you use RStudio (www.rstudio.org) you can find good initial start values by interactive plot using manipulate() function. Look the simple code below. age - 1:20 k - 3; b0 - -5; b1 - .5 y - k*exp(b0+b1*age)/(1+exp(b0+b1*age))+rnorm(age,0,0.1) plot(y~age) start - list() require(manipulate) manipulate( { plot(y~age) k - kk; b0 - b00; b1 - b10 curve(k*exp(b0+b1*x)/(1+exp(b0+b1*x)), add=TRUE) start - list(k=k, b0=b0, b1=b1) }, kk=slider(0,5,step=0.1,initial=2.7), b00=slider(-5,0,step=0.1,initial=-2.5), b10=slider(0,1,step=0.1,initial=0.5)) start n0 - nls(y~k*exp(b0+b1*age)/(1+exp(b0+b1*age)), start=start) summary(n0) If you don't use RStudio you can follow the procedures in this blog (that uses gWidgetsRGtk2). http://ridiculas.wordpress.com/2011/04/09/metodo-grafico-interativo-para-valores-iniciais-em-regressao-nao-linear/ Bests. Walmes. == Walmes Marques Zeviani LEG (Laboratório de Estatística e Geoinformação, 25.450418 S, 49.231759 W) Departamento de Estatística - Universidade Federal do Paraná fone: (+55) 41 3361 3573 VoIP: (3361 3600) 1053 1173 e-mail: wal...@ufpr.br twitter: @walmeszeviani homepage: http://www.leg.ufpr.br/~walmes linux user number: 531218 == On Mon, Jun 6, 2011 at 4:20 AM, Rubén Roa r...@azti.es wrote: Write the growth formula in an R script. Define initial par values. Input the size and age data. Plot the size and age data as points. Plot the growth model with the initial par values as a line. Play with the initial par values until you see a good agreement between the model (the line) and the data (the points). Optimise. Re-plot. Plot a residual histogram. Plot a residual scatterplot. Plot a Q-Q residual plot. HTH Rubén Dr. Rubén Roa-Ureta AZTI - Tecnalia / Marine Research Unit Txatxarramendi Ugartea z/g 48395 Sukarrieta (Bizkaia) SPAIN -Mensaje original- De: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] En nombre de Renalda Enviado el: sábado, 04 de junio de 2011 6:17 Para: r-help@r-project.org Asunto: [R] logistic growth model I want to Fit a logistic growth model for y = k *eb0+b1(age)/1 + eb0+b1(age), can some one help on how to get the initial coefficients b0 and b1? I need to estimate in order to do the regression analysis. When I run using b0=0.5 and b1=3.4818, I get the following error 397443.8 : 0.5 3.4818 Error in nls(Height ~ k * exp(b1 + b2 * Age)/(1 + exp(b1 + b2 * Age)), : singular gradient please tell me what is wrong with my initials values, and how to get the initial values __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How to assess the accuracy of fitted logistic regression using glm
Hi, I am trying glm with family = binomial to do binary logistic regression, but how can I assess the accuracy of the fitted model, the summary method can print a lot of information about the returned object, such as coefficients, because statistics is not my speciality, so can you share some rule of thumb to exam the fitted model from the practical perspective. Regards, Xiaobo Gu __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] simulation
Dear ALL I want to simulate data from Multivariate normal distribution. GE.N-mvrnorm(25,mu,S) S -matrix(rep(0,1),nrow=100) for( i in 1:100){sigma-runif(100,0.1,10);S [i,i]=sigma[i];mu-runif(100,0,10)} for (i in 1:20){for (j in 1:20){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 21:40){for (j in 21:40){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 41:60){for (j in 41:60){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 61:80){for (j in 61:80){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 81:100){for (j in 81:100){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} How should I do when S is not positive definite matrix? I saw this error: 'Sigma' is not positive definite. best regards, Sara [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] qplot fill and colour not working as expected
I am just learning to use qplot and can't get the fill/colour to work. Below is the R code for a scatter plot and bar graph. library(ggplot2) x-c(1,2,3,4,5,6,7) y-c(1,2,3,2,5,6,3) qplot(x,y, main=Scatter Plot Test, xlab=X Label Test, ylab=Y Label Test, colour=blue)z-c(van, van, van, car, car, truck, truck, truck, truck, van, van) qplot(z, main=Bar Graph Test, ylab=Vehichle Count, xlab=Vehicle Category, fill=blue) If I set fill=z, then I can get something different, i.e. three preselected colors are used, but what if I want all three bars to just be blue instead of the default? I am having the same issue if I try to use colour. How do I change the default color to blue instead of black. Second question what exactly is the difference between colour and fill? -- View this message in context: http://r.789695.n4.nabble.com/qplot-fill-and-colour-not-working-as-expected-tp3576949p3576949.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Wireframe, custom x-axis values
Hi, Im plotting some data with wireframe() like so: wireframe(result ~ u * r, myData, scales=list(arrows=FALSE)) However, I would really like to display something different for the displayed values of u rather than the actual values. This is because my u-values are a sequence of quantiles of myData, and I would like to display the quantiles used (e.g. 0.8 0.85 0.9 0.95) instead of the actual values of these quantiles, since this is easier to relate to for a viewer. This information is accessible in myData in a variable, qnt. I've tried meddling around with axis, label and at in scales=list(), but i've been unable to make it happen. Can anyone shed some light? Preferably in a short, generic example. Thanks in advance, Thor [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Not missing at random
Hello!  I would like to sample 30 % of cases (with at least 1 value lower than 3 - in the row) and among them I want to set all values lower than 3 (within selected cases) as NA (NMAR- Not missing at random). I managed to sample cases, but I donât know how to set values (lower than 3) as NA.  R code:  x - matrix(c(1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,3,3,3,4), nrow = 7, ncol=7, byrow=TRUE) matrix  pMiss - 30    percent of missing values  N - dim(x)[1]  number of cases  candidate-which(x[,1]3 | x[,2]3 | x[,3]3 | x[,4]3 | x[,5]3 | x[,6]3 | x[,7]3)   I want to sample all cases with at least 1 value lower than 3, so I have to find candidates  idMiss - sample(candidate, N * p / 100)   I sampled cases  Now I'd like to set all values among sampled cases as NA.  Any suggestion?  Thanks, Blaž [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problem in R documentation
I am not able to run Dickey-Fuller test. adf.test() function is not working. It is showing 'Error: could not find function adf.test Can any tell how to call time series library? -- Siddharth Arun, 4th Year Undergraduate student Industrial Engineering and Management, IIT Kharagpur [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Negating two identical characters with regular expressions
Hello Michael, try strsplit(aa-bb-cc dd, -{2,}) . This function returns an array with all the strings separated by multiple dashes (at least two). Alternatively if you want the first string only try this: sub((.*?)--.*, \\1, aa-bbcc dd) (note the reluctant quantifier *? ) Hope this helps Marco On Sun, Jun 5, 2011 at 9:59 PM, Michael Young michaeltyo...@gmail.comwrote: Hello all, Let's say I have a character string Race-ethnicity-coding information I want to extract all text before the multiple dashes, including the word ethnicity. I wrote a handy function to extract the first matched text: grepcut - function(pattern,x){ start.and.length - regexpr(pattern,x) substring(x,start.and.length,start.and.length +attr(start.and.length,match.length)-1)} grepcut(^[^-]+,Race-ethnicity-coding information) The above grepcut, of course, returns only the string Race What I really want is a to create a class of two dashes in a row and then negate that. Is it possible to create a class of repeated characters? If so, it might be further complicated that - is a special character in brackets and can only go first or last. Can anyone help me out? Thanks, Michael Young [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Ogni tempo ha il suo fascismo. A questo si arriva in molti modi, non necessariamente col terrore dell'intimidazione poliziesca, ma anche negando o distorcendo l'informazione, inquinando la giustizia, paralizzando la scuola, diffondendo in molti sottili modi la nostalgia per un mondo in cui regnava sovrano l'ordine. (Primo Levi, 1974) [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] How to assess the accuracy of fitted logistic regression using glm
On Mon, 6 Jun 2011, Xiaobo Gu wrote: Hi, I am trying glm with family = binomial to do binary logistic regression, but how can I assess the accuracy of the fitted model, the summary method can print a lot of information about the returned object, such as coefficients, because statistics is not my speciality, so can you share some rule of thumb to exam the fitted model from the practical perspective. It depends entirely on why you did the fit. People have written whole books on assessing the performance of classification procedures such as binary logistic regression. For example, the residual deviance is closely related to log-probability scoring: for some purposes that is a good performance measure, for others (e.g. when you are going to threshold the predicted probabilities) it can be very misleading. In short, you need statistical advice, not R advice (the purpose of this list). Regards, Xiaobo Gu __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, rip...@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] RCurl and kerberos
Dear list, I would like to call a Kerberos-authenticated web-service from within R. Curl can do it: $ curl --negotiate -u : http://my.web.service/; so I would expect that RCurl also has the capability, but I have not been able to find the correct options to set. listCurlOptions() does not return anything with negotiate, and searching the source of RCurl, the only thing I found was ./RCurl/R/curlInfo.S:names(CurlFeatureBits) = c(ipv6, kerberos4, ssl, libz, ntlm, gssnegotiate, but e.g. getURL(http://my.web.service,.opts=curlOptions(username=:,httpauth=gssnegotiate)) does not work. Does anybody know if RCurl or another package is able to negotiate the GSS api or otherwise access a Kerberos enabled webservice and how? Testing was done in R 2.12.1 on Ubuntu 10.04.1 LTS / Lucid. Thanks in advance, Thomas. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Problem in R documentation
What operating system are you using? What version of R are you using? How did you install the package in question? Did the installation process give any error messages? Did you load the package before trying to use it? What package are you trying to load - there is no time series package. Do you mean timeSeries? Or tseries? Or some other package entirely? On Mon, Jun 6, 2011 at 2:41 AM, siddharth arun sid.aru...@gmail.com wrote: I am not able to run Dickey-Fuller test. adf.test() function is not working. It is showing 'Error: could not find function adf.test Can any tell how to call time series library? -- -- Sarah Goslee http://www.functionaldiversity.org __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] adding an ellipse to a PCA plot
Hi, I created a principal component plot using the first two principal components. I used the function princomp() to calculate the scores. now, I would like to superimpose an ellipse representing the center and the 95% confidence interval of a series of points in my plot (as to illustrate the grouping of my samples). I looked at the ellipse() function in the ellipse package but can't get it to work. the princomp()-function gives me the scores of each point, so I can calculate the mean and the 95%-CI, but I can't integrate this into the ellipse()-function). Is there a better way of doing this or can someone help me figure out this function? best regards, Lukas __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Problem in R documentation
Hi Siddharth, adf.test() is part of the tseries package, so you need to download and install it before using that function. Try the following and let us now what you get: install.packages('tseries') require(tseries) ?adf.test HTH, Jorge On Mon, Jun 6, 2011 at 2:41 AM, siddharth arun wrote: I am not able to run Dickey-Fuller test. adf.test() function is not working. It is showing 'Error: could not find function adf.test Can any tell how to call time series library? -- Siddharth Arun, 4th Year Undergraduate student Industrial Engineering and Management, IIT Kharagpur [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Lasso for k-subset regression
Hi, On Sun, Jun 5, 2011 at 9:12 PM, Dae-Jin Lee lee.dae...@gmail.com wrote: Dear R-users I'm trying to use lasso in lars package for subset regression, I have a large matrix of size 1000x100 and my aim is to select a subset k of the 100 variables. Is there any way in lars to fix the number k (i.e. to select the best 10 variables) library(lars) aa=lars(X,Y,type=lasso,max.steps=200) plot(aa,plottype=Cp) aa$RSS which.min(aa$RSS) round(aa$beta,2) aa$beta[which.min(aa$RSS),] # find which coefficients minimizes the RSS lasso.ind=which((as.vector((aa$beta[which.min(aa$RSS),])))0) # index of variables print(lasso.ind) # this usually gives more than 10 variables (also depends on the max.steps in lars) First off: I'd suggest using the glmnet package instead of lars. Setting its `alpha` parameter to 1 will give you the lasso, but you can also play w/ different values of alpha to see if an elasticnet-type penalty would be better. Now that you are using glmnet, check its `dfmax` and `pmax` arguments. HTH, -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology | Memorial Sloan-Kettering Cancer Center | Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] qplot fill and colour not working as expected
Hi, On Mon, Jun 6, 2011 at 9:49 AM, wwreith reith_will...@bah.com wrote: I am just learning to use qplot and can't get the fill/colour to work. Below is the R code for a scatter plot and bar graph. library(ggplot2) x-c(1,2,3,4,5,6,7) y-c(1,2,3,2,5,6,3) qplot(x,y, main=Scatter Plot Test, xlab=X Label Test, ylab=Y Label Test, colour=blue)z-c(van, van, van, car, car, truck, truck, truck, truck, van, van) qplot(z, main=Bar Graph Test, ylab=Vehichle Count, xlab=Vehicle Category, fill=blue) If I set fill=z, then I can get something different, i.e. three preselected colors are used, but what if I want all three bars to just be blue instead of the default? I am having the same issue if I try to use colour. How do I change the default color to blue instead of black. Use 'fill=I(blue)' or 'colour=I(blue)'. Second question what exactly is the difference between colour and fill? As near as I can tell, colour applies to lines and points, fill applies to the interier of rectangles and other areas. Best, Ista -- View this message in context: http://r.789695.n4.nabble.com/qplot-fill-and-colour-not-working-as-expected-tp3576949p3576949.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Ista Zahn Graduate student University of Rochester Department of Clinical and Social Psychology http://yourpsyche.org __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] qplot fill and colour not working as expected
One difference between colour and fill can be demonstrated by specifying both with different values. In cases where a polygon is filled, colour specifies border line color. On Mon, Jun 6, 2011 at 9:49 AM, wwreith reith_will...@bah.com wrote: I am just learning to use qplot and can't get the fill/colour to work. Below is the R code for a scatter plot and bar graph. library(ggplot2) x-c(1,2,3,4,5,6,7) y-c(1,2,3,2,5,6,3) qplot(x,y, main=Scatter Plot Test, xlab=X Label Test, ylab=Y Label Test, colour=blue)z-c(van, van, van, car, car, truck, truck, truck, truck, van, van) qplot(z, main=Bar Graph Test, ylab=Vehichle Count, xlab=Vehicle Category, fill=blue) If I set fill=z, then I can get something different, i.e. three preselected colors are used, but what if I want all three bars to just be blue instead of the default? I am having the same issue if I try to use colour. How do I change the default color to blue instead of black. Second question what exactly is the difference between colour and fill? -- View this message in context: http://r.789695.n4.nabble.com/qplot-fill-and-colour-not-working-as-expected-tp3576949p3576949.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- === Jon Daily Technician === #!/usr/bin/env outside # It's great, trust me. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] shading in overlap between two ranges
This worked perfectly. An example graphic is located here: ftp://ftp.sfwmd.gov/pub/ggraves/ribbon.bmp -Original Message- From: Dennis Murphy [mailto:djmu...@gmail.com] Sent: Thursday, June 02, 2011 12:11 PM To: Graves, Gregory Cc: r-help@r-project.org; Kemp, Susan K SAJ; patrick_pi...@fws.gov Subject: Re: [R] shading in overlap between two ranges Hi: Here's one approach using geom_ribbon() in ggplot2 - the 'overlap' is the change in color where the two ribbons intersect. Using your example data with the same names and the 'one.month' variable removed, library(ggplot2) ggplot() + geom_ribbon(data = target, aes(x = i.value, ymin = X25, ymax = X75, fill = 'Target'), alpha = 0.4) + geom_ribbon(data = observed, aes(x = i.value, ymin = X25, ymax = X75, fill = 'Observed'), alpha = 0.4) + scale_fill_manual(, c('Target' = 'blue', 'Observed' = 'orange')) + opts(legend.position = c(0.88, 0.85), legend.background = theme_rect(colour = 'transparent'), legend.text = theme_text(size = 12)) + labs(x = 'Month', y = 'Value') There is a separate geom_ribbon() for each of target and observed. A factor variable for fill color is generated on the fly with colors specified in scale_fill_manual(). The opts() reposition the legend inside the graphics region (the values represent proportions of the total graphics area in each direction), make the legend background transparent and slightly increase the size of the legend labels (default size = 10 in theme_text). Alpha transparency is used so that the overlap creates a blend of the two colors; without it, one overwrites the other. HTH, Dennis On Thu, Jun 2, 2011 at 8:04 AM, Graves, Gregory ggra...@sfwmd.gov wrote: I have 2 datafiles 'target' and 'observed' as shown below (I will gladly email these 2 small files to whomever). X25. And X75. Indicate the value of 25th and 75th-percentile of the target ('what should be') and the observed ('what is'). The i.value is simply the month. target X i.value X25. X75. 1 one.month 1 10.845225 17.87237 2 one.month 2 12.235813 19.74490 3 one.month 3 14.611749 23.44810 4 one.month 4 17.529332 28.09647 5 one.month 5 19.458738 30.56936 6 one.month 6 15.264505 28.29333 7 one.month 7 12.370369 23.35455 8 one.month 8 12.471224 21.82794 9 one.month 9 9.716685 17.28762 10 one.month 10 6.470568 12.49830 11 one.month 11 6.180560 14.24961 12 one.month 12 9.673738 15.79208 observed X i.value X25. X75. 1 one.month 1 19.81000 27.63500 2 one.month 2 23.64062 30.09125 3 one.month 3 26.04865 35.99104 4 one.month 4 32.02625 41.50958 5 one.month 5 34.74479 47.75958 6 one.month 6 37.48885 46.56448 7 one.month 7 30.06740 40.10146 8 one.month 8 26.14917 39.49458 9 one.month 9 14.12521 32.39406 10 one.month 10 11.04125 23.55479 11 one.month 11 13.14917 23.56833 12 one.month 12 17.17938 27.02458 The following plots 4 lines on one graph. The area between the two red lines represents the target 'zone', and the area between the two black lines is the observed 'zone'. with(target, plot(X25.~i.value,ylim=c(0,55),type='l',col='red')) par(new=T) with(target, plot(X75.~i.value,ylim=c(0,55),type='l',col='red')) par(new=T) with(observed, plot(X25.~i.value,ylim=c(0,55),type='l')) par(new=T) with(observed, plot(X75.~i.value,ylim=c(0,55),type='l')) par(new=F) Ideally, the target and the observed should overlap in every month - they don't. The desire is to visually accentuate the amount of overlap by shading in the area where these two zones overlap. How would you do that? Note, that in some of these characterizations, the overlap wanders in and out [I already have routines that calculate the percent of overlap, but I have been requested to find a way to shade the overlap.] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] adding an ellipse to a PCA plot
Dear Lukas, You might try the dataEllipse() function in the car package. I hope this helps, John -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Lukas Baitsch Sent: June-06-11 10:33 AM To: r-help@r-project.org Subject: [R] adding an ellipse to a PCA plot Hi, I created a principal component plot using the first two principal components. I used the function princomp() to calculate the scores. now, I would like to superimpose an ellipse representing the center and the 95% confidence interval of a series of points in my plot (as to illustrate the grouping of my samples). I looked at the ellipse() function in the ellipse package but can't get it to work. the princomp()-function gives me the scores of each point, so I can calculate the mean and the 95%-CI, but I can't integrate this into the ellipse()-function). Is there a better way of doing this or can someone help me figure out this function? best regards, Lukas __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test cases (data) for data based modeling
Hello all, I'm working mostly with machine learning code in R and looking for a structured way to check if my code is working properly. For example if I train a classifier on some data. How do I know if the good / bad results are related to the data are not just an programming error that I introduced somewhere. results are to good: I might have used some part of the test data for training results are to bad: could have any reason I know that I can in principle generate data containing no information at all or pure information to benchmark my code but is there a more elaborate or easyer way to that? I guess what I'm basically looking for is some kind of unit testing framework to generate test data for machine learning tasks, I read about the package RUnit but don't really know how to proceed from there. Any ideas? How do you test your data analysis code? best regards, Immanuel __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How can I write methods for 'as()'?
Dear list, I wonder how to write methods for the function 'as' in the sense that I can call 'as(object, Class, strict=TRUE, ext)' and let method dispatch figure out the correct method. AFAIU, there is a difference between, e.g. 'as.data.frame' and the methods of 'as()' as stated above since the former depends on arg 'x' instead of 'object', 'Class' etc.? methods(as) as.data.frame I have to admit that I'm not really familiar with the S3 style of defining methods as I have been coding in S4 a lot, but my first attempt was to write something like this: as.myClass - function(x, ...){ if(is(x, data.frame){ x - as.list(x) } if(is(x, character){ x - as.list(x) } ... out - getRefClass(myClass)$new(X=x) return(out) } But that way I'd have to explicitly call 'as.myClass(x)' whereas I'd simply like to type 'as(x, myClass)'. Also, how is it possible to have method dispatch recognize two signature arguments in S3? I.e., how can I define something like 'as.data.frame.character' in order to have explicit sub methods for all the data types of 'x' so I wouldn't have to process them all in the definition of 'as.myClass' as I did above? Thanks for your help, Janko -- *Janko Thyson* janko.thy...@googlemail.com mailto:janko.thy...@googlemail.com Jesuitenstraße 3 D-85049 Ingolstadt Mobile: +49 (0)176 83294257 This e-mail and any attachment is for authorized use by the intended recipient(s) only. It may contain proprietary material, confidential information and/or be subject to legal privilege. It should not be copied, disclosed to, retained or used by any other party. If you are not an intended recipient then please promptly delete this e-mail and any attachment and all copies and inform the sender. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] adding an ellipse to a PCA plot
Dear John, Thanks, this actually works just fine! much easier than the ellipse-package! Lukas On Monday, June 6, 2011, John Fox j...@mcmaster.ca wrote: Dear Lukas, You might try the dataEllipse() function in the car package. I hope this helps, John -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Lukas Baitsch Sent: June-06-11 10:33 AM To: r-help@r-project.org Subject: [R] adding an ellipse to a PCA plot Hi, I created a principal component plot using the first two principal components. I used the function princomp() to calculate the scores. now, I would like to superimpose an ellipse representing the center and the 95% confidence interval of a series of points in my plot (as to illustrate the grouping of my samples). I looked at the ellipse() function in the ellipse package but can't get it to work. the princomp()-function gives me the scores of each point, so I can calculate the mean and the 95%-CI, but I can't integrate this into the ellipse()-function). Is there a better way of doing this or can someone help me figure out this function? best regards, Lukas __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] How can I write methods for 'as()'?
Somehow I don't see my own postings in the list, so sorry for replying to my own message and not the one that went out to the list. I got a little further and I think I found exactly the thing that is bothering me: how to get extended method dispatch going in 'setAs()': setRefClass(A, fields=list(X=numeric)) setRefClass(B, contains=A, fields=list(Y=character)) setAs(from=numeric, to=A, def=function(from,to){ out - getRefClass(to)$new(X=from) return(out) } ) a - as(1:5, A) a$X b - as(1:5, B) My problem is the last statement (b - as(1:5, B) which fails. I want to get around having to write new 'setAs' methods for all classes extending class 'A'. If 'B' inherits from 'A', shouldn't it then be possible to tell 'setAs' to look for the next suitable method, i.e. the method defined for 'A'? I tried 'NextMethod()' inside the body of 'setAs' but that didn't work out. Thanks a lot, Janko On 06.06.2011 17:15, Janko Thyson wrote: Dear list, I wonder how to write methods for the function 'as' in the sense that I can call 'as(object, Class, strict=TRUE, ext)' and let method dispatch figure out the correct method. AFAIU, there is a difference between, e.g. 'as.data.frame' and the methods of 'as()' as stated above since the former depends on arg 'x' instead of 'object', 'Class' etc.? methods(as) as.data.frame I have to admit that I'm not really familiar with the S3 style of defining methods as I have been coding in S4 a lot, but my first attempt was to write something like this: as.myClass - function(x, ...){ if(is(x, data.frame){ x - as.list(x) } if(is(x, character){ x - as.list(x) } ... out - getRefClass(myClass)$new(X=x) return(out) } But that way I'd have to explicitly call 'as.myClass(x)' whereas I'd simply like to type 'as(x, myClass)'. Also, how is it possible to have method dispatch recognize two signature arguments in S3? I.e., how can I define something like 'as.data.frame.character' in order to have explicit sub methods for all the data types of 'x' so I wouldn't have to process them all in the definition of 'as.myClass' as I did above? Thanks for your help, Janko [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] simulation
On Mon, Jun 06, 2011 at 04:50:57PM +1000, Stat Consult wrote: Dear ALL I want to simulate data from Multivariate normal distribution. GE.N-mvrnorm(25,mu,S) S -matrix(rep(0,1),nrow=100) for( i in 1:100){sigma-runif(100,0.1,10);S [i,i]=sigma[i];mu-runif(100,0,10)} for (i in 1:20){for (j in 1:20){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 21:40){for (j in 21:40){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 41:60){for (j in 41:60){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 61:80){for (j in 61:80){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} for (i in 81:100){for (j in 81:100){if (i != j){S [i,j]=0.3*sigma[i]*sigma[j]}}} How should I do when S is not positive definite matrix? I saw this error: 'Sigma' is not positive definite. Hello. I am not sure, how the matrix is created. Should the command sigma-runif(100,0.1,10) be indeed inside the loop over i? I suspect that no, since otherwise, only the vector sigma used for S[100, 100] goes to the remaining part of the construction. The matrix is block diagonal. So, the corresponding distribution can be build from parts corresponding to the blocks generated independently. Let me look at the first block assuming that sigma is generated only once. The first block may be obtained also as B - 0.3*outer(sigma[1:20], sigma[1:20]) diag(B) - sigma[1:20] The result may have negative eigenvalues. For example, if all components in sigma[1:20] are 4, which is in the range used for sigma, then we have a matrix, whose diagonal elements are 4 and nondiagonal elements are 0.3*4^2 = 4.8 4. This matrix has negative eigenvalues, so it is not a covariance matrix. Is the construction of the matrix, which you sent, correct? Petr Savicky. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] How can I write methods for 'as()'?
Okay, I found something that is working, but it looks and feels pretty awkward as the method def and method lookup takes place in one function ;-) setRefClass(A, fields=list(X=numeric)) setRefClass(B, contains=A, fields=list(Y=character)) mySetAs - function( from, to ){ if(!existsMethod(f=coerce, signature=c(from=class(from), to=to))){ setAs(from=class(from), to=to, def=function(from, to){ out - getRefClass(to)$new(X=from) return(out) } ) } mthd - selectMethod(coerce, signature=c(from=class(from), to=to), useInherited= c(from=TRUE, to=TRUE)) out - mthd(from=from, to=to) return(out) } a - mySetAs(from=1:5, to=A) a$X b - mySetAs(from=1:5, to=B) b$X I'm sure there are much better ways. I'd appreciate any comments whatsoever. Best regards, Janko On 06.06.2011 17:46, Janko Thyson wrote: Somehow I don't see my own postings in the list, so sorry for replying to my own message and not the one that went out to the list. I got a little further and I think I found exactly the thing that is bothering me: how to get extended method dispatch going in 'setAs()': setRefClass(A, fields=list(X=numeric)) setRefClass(B, contains=A, fields=list(Y=character)) setAs(from=numeric, to=A, def=function(from,to){ out - getRefClass(to)$new(X=from) return(out) } ) a - as(1:5, A) a$X b - as(1:5, B) My problem is the last statement (b - as(1:5, B) which fails. I want to get around having to write new 'setAs' methods for all classes extending class 'A'. If 'B' inherits from 'A', shouldn't it then be possible to tell 'setAs' to look for the next suitable method, i.e. the method defined for 'A'? I tried 'NextMethod()' inside the body of 'setAs' but that didn't work out. Thanks a lot, Janko On 06.06.2011 17:15, Janko Thyson wrote: Dear list, I wonder how to write methods for the function 'as' in the sense that I can call 'as(object, Class, strict=TRUE, ext)' and let method dispatch figure out the correct method. AFAIU, there is a difference between, e.g. 'as.data.frame' and the methods of 'as()' as stated above since the former depends on arg 'x' instead of 'object', 'Class' etc.? methods(as) as.data.frame I have to admit that I'm not really familiar with the S3 style of defining methods as I have been coding in S4 a lot, but my first attempt was to write something like this: as.myClass - function(x, ...){ if(is(x, data.frame){ x - as.list(x) } if(is(x, character){ x - as.list(x) } ... out - getRefClass(myClass)$new(X=x) return(out) } But that way I'd have to explicitly call 'as.myClass(x)' whereas I'd simply like to type 'as(x, myClass)'. Also, how is it possible to have method dispatch recognize two signature arguments in S3? I.e., how can I define something like 'as.data.frame.character' in order to have explicit sub methods for all the data types of 'x' so I wouldn't have to process them all in the definition of 'as.myClass' as I did above? Thanks for your help, Janko [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Wireframe, custom x-axis values
On 2011-06-06 06:33, Rbjørn Nicolaisen wrote: Hi, Im plotting some data with wireframe() like so: wireframe(result ~ u * r, myData, scales=list(arrows=FALSE)) However, I would really like to display something different for the displayed values of u rather than the actual values. This is because my u-values are a sequence of quantiles of myData, and I would like to display the quantiles used (e.g. 0.8 0.85 0.9 0.95) instead of the actual values of these quantiles, since this is easier to relate to for a viewer. This information is accessible in myData in a variable, qnt. I've tried meddling around with axis, label and at in scales=list(), but i've been unable to make it happen. Can anyone shed some light? Preferably in a short, generic example. Here is a slight modification of the second example in help('wireframe'): wireframe(z ~ x * y, data = g, groups = gr, scales = list(arrows = FALSE, x = list(at = c(2, 5, 10)), y = list(at = c(6, 10, 14), lab = c('A', 'BBB', 'C')) )) Peter Ehlers Thanks in advance, Thor [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Seeking help to define method for show() for an S4 object
Dear all, I have created a new S4 class with name MyClass. Please see below for it's definition. Now I want to create a method for the show() function for this class. In defining this method, what I want is, once user would like see an object of this class some values will be displayed. Then R will ask for **press Enter**. Once user presses enter then, remaining values will be displayed. In the following example, I try to explain this concept. setClass(MyClass, sealed = FALSE, representation(Slot1 = vector, Slot2 = vector)) [1] MyClass setMethod(show, MyClass, definition = function(x) { + cat(These are the values. Please press enter to see the values.\n) + ### User will presss the Enter ## + ### Then only following figures will be visible # + cat(x@Slot1) +} ) [1] show Warning message: For function show, signature MyClass: argument in method definition changed from (x) to (object) new(MyClass, Slot1 = 1:3, Slot2 = 4:7) These are the values. Please press enter to see the values. 1 2 3 Can somebody guide me how I can achieve that? Thanks, [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Seeking help to define method for show() for an S4 object
Hi, Take a look at ?readLines Also, you should use function(object) not function(x) because of how the generic is defined. Below is an example. Cheers, Josh setClass(MyClass, sealed = FALSE, representation( Slot1 = vector, Slot2 = vector)) setMethod(f = show, signature = MyClass, definition = function(object) { cat(These are the values. Please press enter to see the values.\n) readLines(n = 1) cat(object@Slot1) }) x - new(MyClass, Slot1 = 1:3, Slot2 = 4:7) x On Mon, Jun 6, 2011 at 10:19 AM, Bogaso Christofer bogaso.christo...@gmail.com wrote: Dear all, I have created a new S4 class with name MyClass. Please see below for it's definition. Now I want to create a method for the show() function for this class. In defining this method, what I want is, once user would like see an object of this class some values will be displayed. Then R will ask for **press Enter**. Once user presses enter then, remaining values will be displayed. In the following example, I try to explain this concept. setClass(MyClass, sealed = FALSE, representation(Slot1 = vector, Slot2 = vector)) [1] MyClass setMethod(show, MyClass, definition = function(x) { + cat(These are the values. Please press enter to see the values.\n) + ### User will presss the Enter ## + ### Then only following figures will be visible # + cat(x@Slot1) + } ) [1] show Warning message: For function show, signature MyClass: argument in method definition changed from (x) to (object) new(MyClass, Slot1 = 1:3, Slot2 = 4:7) These are the values. Please press enter to see the values. 1 2 3 Can somebody guide me how I can achieve that? Thanks, [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Joshua Wiley Ph.D. Student, Health Psychology University of California, Los Angeles http://www.joshuawiley.com/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Question with RExcel
Dear all, I’m doing some simulation studies in order to compare the estimates (and estimated standard deviations) from the ARMA(2,1) Model with an estimator that I’ve constructed. For carrying out the simulations I created a VBA project within Excel. Now, I’m using the RExcel tool for running the R commands in the VBA project. I run 2500 simulation using the “arima” function from R and it worked! Nevertheless, the constant was badly estimated. So, I decided to use the “arma” function from R, and the parameters are now well estimated. However, I cannot run the 2500 simulations. It can only do 46 simulations! I’ve already tried to run the program in another computer, but I’ve got the same problem. Do you have any suggestions? Thanks for your attention. Helena Mouriño. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] generating random covariance matrices (with a uniform distribution of correlations)
Thank you very much, this does help quite a bit. Ned From: Petr Savicky savicky_at_praha1.ff.cuni.cz Date: Sat, 04 Jun 2011 11:44:52 +0200 On Fri, Jun 03, 2011 at 01:54:33PM -0700, Ned Dochtermann wrote: Petr, This is the code I used for your suggestion: k-6;kk-(k*(k-1))/2 x-matrix(0,5000,kk) for(i in 1:5000){ A.1-matrix(0,k,k) rs-runif(kk,min=-1,max=1) A.1[lower.tri(A.1)]-rs A.1[upper.tri(A.1)]-t(A.1)[upper.tri(A.1)] cors.i-diag(k) t-.001-min(Re(eigen(A.1)$values)) new.cor-cov2cor(A.1+(t*cors.i)) x[i,]-new.cor[lower.tri(new.cor)]} hist(c(x)); max(c(x)); median(c(x)) This, unfortunately, does not maintain the desired distribution of correlations. Hello. On the contrary to what i thought originally, there are solutions also for the case of the correlation matrix. The first solution creates a singular correlation matrix (of rank 3), but the nondiagonal entries have exactly the uniform distribution on [-1, 1], since the scalar product of two independent uniformly distributed unit vectors in R^3 has the uniform distribution on [-1, 1]. x - matrix(rnorm(18), nrow=6, ncol=3) x - x/sqrt(rowSums(x^2)) a - x %*% t(x) The next solution produces a correlation matrix of full rank, whose non-diagonal entries have distribution very close to the uniform on [-1, 1]. KS test finds a difference only with sample size more than 50'000. w - c(0.01459422, 0.01830718, 0.04066405, 0.50148488, 0.60330865, 0.61832829) x - matrix(rnorm(36), nrow=6, ncol=6) %*% diag(w) x - x/sqrt(rowSums(x^2)) a - x %*% t(x) Hope this helps. Petr Savicky. -Original Message- From: Ned Dochtermann [mailto:ned.dochterm...@gmail.com] Sent: Friday, June 03, 2011 1:55 PM To: 'r-help@r-project.org'; 'savi...@praha1.fff.cuni.cz' Subject: Re: [R] generating random covariance matrices (with a uniform distribution of correlations) Petr, This is the code I used for your suggestion: k-6;kk-(k*(k-1))/2 x-matrix(0,5000,kk) for(i in 1:5000){ A.1-matrix(0,k,k) rs-runif(kk,min=-1,max=1) A.1[lower.tri(A.1)]-rs A.1[upper.tri(A.1)]-t(A.1)[upper.tri(A.1)] cors.i-diag(k) t-.001-min(Re(eigen(A.1)$values)) new.cor-cov2cor(A.1+(t*cors.i)) x[i,]-new.cor[lower.tri(new.cor)]} hist(c(x)); max(c(x)); median(c(x)) This, unfortunately, does not maintain the desired distribution of correlations. I did, however, learn some neat coding tricks (that were new for me) along the way. Ned -- On Thu, Jun 02, 2011 at 04:42:59PM -0700, Ned Dochtermann wrote: List members, Via searches I've seen similar discussion of this topic but have not seen resolution of the particular issue I am experiencing. If my search on this topic failed, I apologize for the redundancy. I am attempting to generate random covariance matrices but would like the corresponding correlations to be uniformly distributed between -1 and 1. ... Any recommendations on how to generate the desired covariance matrices would be appreciated. Hello. Let me suggest the following procedure. 1. Generate a symmetric matrix A with the desired distribution of the non-diagonal elements and with zeros on the diagonal. 2. Compute the smallest eigenvalue lambda_1 of A. 3. Replace A by A + t I, where I is the identity matrix and t is a number such that t + lambda_1 0. The resulting matrix will have the same non-diagonal elements as A, but will be positive definite. Petr Savicky. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Question with RExcel
It's hard to see where the problem is from this information. I would suggest subscribing to and asking this question of the RExcel mailing list (accessible from http://rcom.univie.ac.at/) and providing more detail of what you're trying to do, what is going wrong, error messages (is it R or Excel giving the error?) and so on. For all I know, it may well not be an R issue at all but a problem somewhere in your Excel or VBA setup. Best, Ethan On Mon, Jun 6, 2011 at 12:39 PM, Maria Helena Mourino Silva Nunes mhnu...@fc.ul.pt wrote: Dear all, I’m doing some simulation studies in order to compare the estimates (and estimated standard deviations) from the ARMA(2,1) Model with an estimator that I’ve constructed. For carrying out the simulations I created a VBA project within Excel. Now, I’m using the RExcel tool for running the R commands in the VBA project. I run 2500 simulation using the “arima” function from R and it worked! Nevertheless, the constant was badly estimated. So, I decided to use the “arma” function from R, and the parameters are now well estimated. However, I cannot run the 2500 simulations. It can only do 46 simulations! I’ve already tried to run the program in another computer, but I’ve got the same problem. Do you have any suggestions? Thanks for your attention. Helena Mouriño. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Plot many x and y
Dear all could you please plot many x's and y's with one legend per plot? I would like to thank you in advance for your help Best Regards Alex. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Plot many x and y
Hi Alex, You have not given enough details for us to answer your question. Is this something like what you mean? par(mfcol = c(2, 2)) plot(1:10, 1:10, pch = 16) legend(x = 2, y = 8, legend = Test 1, pch = 16) plot(1:10, 1:10, pch = 15) legend(x = 2, y = 8, legend = Test 2, pch = 15) plot(1:10, 1:10, pch = 14) legend(x = 2, y = 8, legend = Test 3, pch = 14) plot(1:10, 1:10, pch = 13) legend(x = 2, y = 8, legend = Test 4, pch = 13) The posting guide ( http://www.R-project.org/posting-guide.html ) provides some helpful tips on how to write a question that will get a good answer. Cheers, Josh On Mon, Jun 6, 2011 at 12:46 PM, Alaios ala...@yahoo.com wrote: Dear all could you please plot many x's and y's with one legend per plot? I would like to thank you in advance for your help Best Regards Alex. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Joshua Wiley Ph.D. Student, Health Psychology University of California, Los Angeles http://www.joshuawiley.com/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Plot many x and y
Hi Alex, could you be a little more specific as to what exactly you mean by plotting many x's and y's with one legend per plot? Please note what appears at the bottom of every R-help mail: PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Following this piece of advice usually increases your chances for a helpful answer. Best, Stephan Am 06.06.2011 21:46, schrieb Alaios: Dear all could you please plot many x's and y's with one legend per plot? I would like to thank you in advance for your help Best Regards Alex. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Merge two columns of a data frame
I have the following data: prefix - c(cheap, budget) roots - c(car insurance, auto insurance) suffix - c(quote, quotes) prefix2 - c(cheap, budget) roots2 - c(car insurance, auto insurance) roots3 - c(car insurance, auto insurance) suffix3 - c(quote, quotes) df1 - expand.grid(prefix, roots, suffix) df2 - expand.grid(prefix2, roots2) df3 - expand.grid(roots3, suffix3) df1; df2; df3 df1, df2, and df3 are seperate data structures with seperate columns for root, prefix, and suffix. Var1 Var2 Var3 1 cheap car insurance quote 2 budget car insurance quote 3 cheap auto insurance quote 4 budget auto insurance quote 5 cheap car insurance quotes 6 budget car insurance quotes 7 cheap auto insurance quotes 8 budget auto insurance quotes Var1 Var2 1 cheap car insurance 2 budget car insurance 3 cheap auto insurance 4 budget auto insurance Var1 Var2 1 car insurance quote 2 auto insurance quote 3 car insurance quotes 4 auto insurance quotes I want to merge df1, df2, and df3, into one data frame column which looks like. Var1 'cheap car insurance quote' 'budget car insurance quote' 'cheap auto insurance quote' 'budget auto insurance quote' 'cheap car insurance quotes' 'budget car insurance quotes' 'cheap auto insurance quotes' 'budget auto insurance quotes' 'cheap car insurance' 'budget car insurance' 'cheap auto insurance' 'budget auto insurance' 'car insurance quote' 'auto insurance quote' 'car insurance quotes' 'auto insurance quotes' Help! WebRep Overall rating [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Problem in R documentation
Thanks Jorge it worked On Mon, Jun 6, 2011 at 8:02 PM, Jorge Ivan Velez jorgeivanve...@gmail.comwrote: Hi Siddharth, adf.test() is part of the tseries package, so you need to download and install it before using that function. Try the following and let us now what you get: install.packages('tseries') require(tseries) ?adf.test HTH, Jorge On Mon, Jun 6, 2011 at 2:41 AM, siddharth arun wrote: I am not able to run Dickey-Fuller test. adf.test() function is not working. It is showing 'Error: could not find function adf.test Can any tell how to call time series library? -- Siddharth Arun, 4th Year Undergraduate student Industrial Engineering and Management, IIT Kharagpur [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Siddharth Arun, 4th Year Undergraduate student Industrial Engineering and Management, IIT Kharagpur [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Denton benchmarking method
Good morning, Does it exist R code to do Denton benchmarking for time series ? Thank you. P. Grandeau __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problem plotting in R under linux (centos)
Dear All, I am running into a slightly odd issue when attempting to produce a png/pdf file in R under centos. The version of Centos is 5.6 with an old version of R 2.12.0 (2010-10-15). My script is roughly: mydata - read.csv(myfile) png(outfile) plot(0,0,type='n',ylim=c(-500,500),xlim=c(0,700),xlab=... lines(mydata$x,mydata$y) dev.off() It produces a plot with the correct limits and labels but the lines command does not seem to always work (I have checked that the data fall within the defined limits!). No error message is reported. The data sets are reasonable large (~6.5 million lines) but not large enough to cause an issue? It runs fine on my windows machine albeit with the latest R build. If I loop over the data set plotting chunks at a time it works just fine. n - floor(nrow(mydata)/100) for(i in 0:n){ lines(mydata[max((i*100),1) : min(((i+1)*100),nrow(mydata)),]) } I have googled and cant find anything helpful. Is it an issue with R, my R installation, Centos etc Thanks in advance, Roger __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Wireframe, custom x-axis values
Im plotting some data with wireframe() like so: wireframe(result ~ u * r, myData, scales=list(arrows=FALSE)) However, I would really like to display something different for the displayed values of u rather than the actual values. This is because my u-values are a sequence of quantiles of myData, and I would like to display the quantiles used (e.g. 0.8 0.85 0.9 0.95) instead of the actual values of these quantiles, since this is easier to relate to for a viewer. This information is accessible in myData in a variable, qnt. I've tried meddling around with axis, label and at in scales=list(), but i've been unable to make it happen. # Right, because you haven't understood that base graphics are different than lattice or ggplot graphics. Suggest you read the help page for xyplot which has material on the names of list elements used. Pay particular attention to the arguments list in the section on scales. at is a proper argument. label is not. You should also look at the list structure of you plot object with str(). #--- Can anyone shed some light? Preferably in a short, generic example. # You are the one responsible for providing examples on rhelp. \# -- David -- View this message in context: http://r.789695.n4.nabble.com/Wireframe-custom-x-axis-values-tp3576963p3577167.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] adding an ellipse to a PCA plot
Hi, I think the easiest way is to use the function plotellipses of the FactoMineR package (but you have to do your PCA with the PCA function included in this package). Alain On 06-Jun-11 16:32, Lukas Baitsch wrote: Hi, I created a principal component plot using the first two principal components. I used the function princomp() to calculate the scores. now, I would like to superimpose an ellipse representing the center and the 95% confidence interval of a series of points in my plot (as to illustrate the grouping of my samples). I looked at the ellipse() function in the ellipse package but can't get it to work. the princomp()-function gives me the scores of each point, so I can calculate the mean and the 95%-CI, but I can't integrate this into the ellipse()-function). Is there a better way of doing this or can someone help me figure out this function? best regards, Lukas __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Alain Guillet Statistician and Computer Scientist SMCS - IMMAQ - Université catholique de Louvain http://www.uclouvain.be/smcs Bureau c.316 Voie du Roman Pays, 20 B-1348 Louvain-la-Neuve Belgium tel: +32 10 47 30 50 Accès: http://www.uclouvain.be/323631.html __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Taking Integral and Optimization using Integrate, Optim and maxNR
Dear All, Hello! I have some questoins in R programming as follows: Question 1- How to take the integral of this function with respect to y, such that x would appear in the output after taking integral. f(x,y)=(0.1766*exp(-exp(y+lnx))*-exp(y+lnx))/(1-exp(-exp(y+lnx))) y in (-6.907,-1.246) It is doable in maple but not in R. At least I could not find the way. p.s: result from maple is: g(x)=dilog*exp(0.001000755564*x)+0.5*ln(exp(0.001000755564*x))^2-dilog*(exp(0.287653*x))-0.5*ln(exp(0.287653*x))^2 Where dilog=integral(log(t)/(1-t)) for t in (1,x) Question 2- Then I want to optimize (maximize) the answer of the above integral for x. Assum x in (0,100) The answer should be something between 26 and 27. What I did is: got answer of integral(f(x,y)) from maple which is g(x), and then applied it in R. The code is as follows: ##In the case n=1 library(stats) require(graphics) integrand=function(t){(log(t))/(1-t)} #dilog=integrate(integrand, lower=1, upper=x) fr - function(x){(integrate(integrand, lower=1, upper=x)$value)*(exp(0.001000755564*x))+0.5*log(exp(0.001000755564*x))^2-(integrate(integrand, lower=1, upper=x)$value)*(exp(0.287653*x))-0.5*log(exp(0.287653*x))^2} optim(20, fr, NULL, method = BFGS) Question 2-1-Default by optim is to minimize a function, how I can use it to maximization? Question 2-2- The above code faced with errors, and did not work. What I guess is there is something wrong with taking integral. The output of integrate function in R is some sort of thing. I had to somehow tell it, just take the value and forget about the others. But I guess it is still something wrong with it. I also tried maxNR, but is didn't work either. Question 2-3- Thoes above are the easiest case of my problem. Assume the case that I have summation of f(x1,y)+f(x2,y)+...+f(x12,y) The article have done it by E04JAF-NAG Fortran Library Routine Document. But I want to do it in R. Thanx all. Cheers, Maryam [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] draw text outside plot boundaries
i'd like to use the text() function to annotate some points, but the labels get cropped, if the point is on the right is there a way to prevent this, and tell the text() function to allow writing outside the boundaries of the current plot? i don't mind if it looks messy and steps on the margin a bit. - erik __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] RPostgreSQL snowfall
Dear expeRts, I'm currently trying to get data from a PostgreSQL database _in parallel_. I tried two methods: * passing the DBI Connection object to the cluster nodes * connecting the the DB on each node (1) The execution of the first method looks like this: result = sfClusterApplyLB(input, fun, dbiCon) and produces an expired PostgreSQLConnection error. (Of course the passed Connection Object is usable at that moment and afterwards!) (2) For the creation of DB connections on every node a function handling the whole connection is sourced into every node. This function works perfectly without snowfall. Calling it with sfClusterEval(dbConnect()) again only expired connection objects are produced. Even if I create the connection 'a line above' the code which is connecting to the DB it doesn't work... Is there a possibility to connect to PostgreSQL using snowfall? -- with kind regards Florian __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] lme, stepAIC, predict: scope and visibility
Hello all, I've run into a problem where I can't run predict.lme on an object simplified via a stepAIC. A similar post has been recorded on this list: https://stat.ethz.ch/pipermail/r-help/2008-May/162047.html but in my case, I'm going to great lengths to not repeat that poster's error and still coming up short. Any advice would be much appreciated. It would seem that, after stepAIC, predict.lme cannot find the training data for the lme object (and why is it even needed?). Here's some example code: foo = function() { x = c(1:20, 1:20) y = c(1:20, 5 + (1:20)) + rnorm(40) s = c(rep(1, 20), rep(2, 20)) library(lattice) xyplot(y~x|s) dframe = data.frame(x, y, s) m = lme(y~x, random=~1|s, method='ML') newdf = data.frame(x=40, s = 2) res = predict(m, newdata=newdf) print(res) m2 = stepAIC(m, k=log(nrow(dframe))) #res2 = predict(m2, newdata=newdf) res2 = eval(substitute(predict(mL, newdata=nL), list(mL=m2, nL=newdf))) print(res2) } foo() 2 45.86875 attr(,label) [1] Predicted values Start: AIC=136.4 y ~ x Error in eval(expr, envir, enclos) : object 'y' not found [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] list demographics
Hi all, I got curious about something, so in proper scientific fashion I obtained some data and analyzed it. Question: what is the female participation in the R-help email list? Data: the most recent list postings, obtained from the website. I took my best shot at classifying the names given in the email header as male/female, but ended with a fair number of unknowns. This dataset had 2797 list messages, in 895 questions. 1501 messages were replies to one of those questions by someone not the original querent. Across all messages, 6.5% were from women, 77.8% from men, 15.7% unknown. For new questions, 11.7% were from women, 61.2% from men, 27.0% unknown. Among responses to other people's questions, 2.6% were from women, 92.3% from men, 5.1% unknown. Nine women answered other people's questions, but only two were what I'd consider active participants, offering more than two answers. (Not divided up by separate questions, so could be several replies in one discussion.) For men, 214 answered questions, and 90 offered more than two answers. (Not divided up by separate questions, so could be several replies in one discussion.) Six active participants were unclassifiable, so even if all of those were female, that would still be only eight women actively participating in the list in this sample. Is the list representative of statisticians? People who use R? People who participate in statistical software email lists? I have no idea, but I found it interesting that there is so little female participation in the list, even asking questions (where you'd expect to see students and new R users), and almost no female participation in answering questions. For those of you who teach, are your classes heavily skewed? Sarah -- Sarah Goslee http://www.functionaldiversity.org __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Not missing at random
Hi Blaz, See below. x - matrix(c(1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,3,3,3,4), nrow = 7, ncol=7, byrow=TRUE) matrix pMiss - 30 percent of missing values N - dim(x)[1] number of cases candidate - which(x[,1]3 | x[,2]3 | x[,3]3 | x[,4]3 | x[,5]3 | x[,6]3 | x[,7]3) I want to sample all cases with at least 1 value lower than 3, so I have to find candidates ## easier to use this ## find all x 3 and return their row and column indices ## select only row indices, and then find unique candidate - unique(which(x 3, arr.ind = TRUE)[, row]) idMiss - sample(candidate, N * pMiss / 100) I sampled cases ## from the subset of x cases that will be missing ## find all that are 3 and set to NA x[idMiss, ][x[idMiss, ] 3] - NA ## If you are going to do this a lot, consider a function nmar - function(x, op = , value = 3, p = 30) { op - get(op) candidate - unique(which(op(x, value), arr.ind = TRUE)[, row]) idMiss - sample(candidate, nrow(x) * p / 100) x[idMiss, ][op(x[idMiss, ], value)] - NA return(x) } nmar(x) ## has the advantage that you can easily change ## p, the cut off value, the operator (e.g., , , =, etc.) Cheers, Josh On Sun, Jun 5, 2011 at 11:17 PM, Blaz Simcic blazsim...@yahoo.com wrote: Hello! I would like to sample 30 % of cases (with at least 1 value lower than 3 - in the row) and among them I want to set all values lower than 3 (within selected cases) as NA (NMAR- Not missing at random). I managed to sample cases, but I don’t know how to set values (lower than 3) as NA. R code: x - matrix(c(1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,3,3,3,4), nrow = 7, ncol=7, byrow=TRUE) matrix pMiss - 30 percent of missing values N - dim(x)[1] number of cases candidate-which(x[,1]3 | x[,2]3 | x[,3]3 | x[,4]3 | x[,5]3 | x[,6]3 | x[,7]3) I want to sample all cases with at least 1 value lower than 3, so I have to find candidates idMiss - sample(candidate, N * p / 100) I sampled cases Now I'd like to set all values among sampled cases as NA. Any suggestion? Thanks, Blaž [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Joshua Wiley Ph.D. Student, Health Psychology University of California, Los Angeles http://www.joshuawiley.com/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Merge two columns of a data frame
Hi Abraham, Just take it step by step. Paste the values together, combine them, and assign them to a data.frame column. Like this perhaps: df.1.2.3 - data.frame(Var1 = c(with(df1, paste(Var1, Var2, Var3)), with(df2, paste(Var1, Var2)), with(df3, paste(Var1, Var2 Best, Ista On Mon, Jun 6, 2011 at 12:22 PM, Abraham Mathew abra...@thisorthat.com wrote: I have the following data: prefix - c(cheap, budget) roots - c(car insurance, auto insurance) suffix - c(quote, quotes) prefix2 - c(cheap, budget) roots2 - c(car insurance, auto insurance) roots3 - c(car insurance, auto insurance) suffix3 - c(quote, quotes) df1 - expand.grid(prefix, roots, suffix) df2 - expand.grid(prefix2, roots2) df3 - expand.grid(roots3, suffix3) df1; df2; df3 df1, df2, and df3 are seperate data structures with seperate columns for root, prefix, and suffix. Var1 Var2 Var3 1 cheap car insurance quote 2 budget car insurance quote 3 cheap auto insurance quote 4 budget auto insurance quote 5 cheap car insurance quotes 6 budget car insurance quotes 7 cheap auto insurance quotes 8 budget auto insurance quotes Var1 Var2 1 cheap car insurance 2 budget car insurance 3 cheap auto insurance 4 budget auto insurance Var1 Var2 1 car insurance quote 2 auto insurance quote 3 car insurance quotes 4 auto insurance quotes I want to merge df1, df2, and df3, into one data frame column which looks like. Var1 'cheap car insurance quote' 'budget car insurance quote' 'cheap auto insurance quote' 'budget auto insurance quote' 'cheap car insurance quotes' 'budget car insurance quotes' 'cheap auto insurance quotes' 'budget auto insurance quotes' 'cheap car insurance' 'budget car insurance' 'cheap auto insurance' 'budget auto insurance' 'car insurance quote' 'auto insurance quote' 'car insurance quotes' 'auto insurance quotes' Help! WebRep Overall rating [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Ista Zahn Graduate student University of Rochester Department of Clinical and Social Psychology http://yourpsyche.org __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Merge two columns of a data frame
Another possibility: dfs - list(df1, df2, df3) df.1.2.3 - as.data.frame(unlist(sapply(dfs, function(x) do.call(paste, x On Mon, Jun 6, 2011 at 2:37 PM, Ista Zahn iz...@psych.rochester.edu wrote: Hi Abraham, Just take it step by step. Paste the values together, combine them, and assign them to a data.frame column. Like this perhaps: df.1.2.3 - data.frame(Var1 = c(with(df1, paste(Var1, Var2, Var3)), with(df2, paste(Var1, Var2)), with(df3, paste(Var1, Var2 Best, Ista On Mon, Jun 6, 2011 at 12:22 PM, Abraham Mathew abra...@thisorthat.com wrote: I have the following data: prefix - c(cheap, budget) roots - c(car insurance, auto insurance) suffix - c(quote, quotes) prefix2 - c(cheap, budget) roots2 - c(car insurance, auto insurance) roots3 - c(car insurance, auto insurance) suffix3 - c(quote, quotes) df1 - expand.grid(prefix, roots, suffix) df2 - expand.grid(prefix2, roots2) df3 - expand.grid(roots3, suffix3) df1; df2; df3 df1, df2, and df3 are seperate data structures with seperate columns for root, prefix, and suffix. Var1 Var2 Var3 1 cheap car insurance quote 2 budget car insurance quote 3 cheap auto insurance quote 4 budget auto insurance quote 5 cheap car insurance quotes 6 budget car insurance quotes 7 cheap auto insurance quotes 8 budget auto insurance quotes Var1 Var2 1 cheap car insurance 2 budget car insurance 3 cheap auto insurance 4 budget auto insurance Var1 Var2 1 car insurance quote 2 auto insurance quote 3 car insurance quotes 4 auto insurance quotes I want to merge df1, df2, and df3, into one data frame column which looks like. Var1 'cheap car insurance quote' 'budget car insurance quote' 'cheap auto insurance quote' 'budget auto insurance quote' 'cheap car insurance quotes' 'budget car insurance quotes' 'cheap auto insurance quotes' 'budget auto insurance quotes' 'cheap car insurance' 'budget car insurance' 'cheap auto insurance' 'budget auto insurance' 'car insurance quote' 'auto insurance quote' 'car insurance quotes' 'auto insurance quotes' Help! WebRep Overall rating [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Ista Zahn Graduate student University of Rochester Department of Clinical and Social Psychology http://yourpsyche.org __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RPostgreSQL snowfall
I don't think you can share dbi connections across different instances of R. just have each of your helper functions open a local connection. or alternatively, load a package on each instance which keeps a dbi connection open. and make sure you bump up your allowed number of connections in pg_conf if you need to. -Whit On Mon, Jun 6, 2011 at 12:40 PM, Florian Endel flor...@endel.at wrote: Dear expeRts, I'm currently trying to get data from a PostgreSQL database _in parallel_. I tried two methods: * passing the DBI Connection object to the cluster nodes * connecting the the DB on each node (1) The execution of the first method looks like this: result = sfClusterApplyLB(input, fun, dbiCon) and produces an expired PostgreSQLConnection error. (Of course the passed Connection Object is usable at that moment and afterwards!) (2) For the creation of DB connections on every node a function handling the whole connection is sourced into every node. This function works perfectly without snowfall. Calling it with sfClusterEval(dbConnect()) again only expired connection objects are produced. Even if I create the connection 'a line above' the code which is connecting to the DB it doesn't work... Is there a possibility to connect to PostgreSQL using snowfall? -- with kind regards Florian __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] draw text outside plot boundaries
On 2011-06-06 08:44, Erik Aronesty wrote: i'd like to use the text() function to annotate some points, but the labels get cropped, if the point is on the right is there a way to prevent this, and tell the text() function to allow writing outside the boundaries of the current plot? Go to ?par and check out the 'xpd' parameter. Peter Ehlers i don't mind if it looks messy and steps on the margin a bit. - erik __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] A Calculation on list object
Hello, I am into some calculation on a list object, therefore requesting the peers if there is any short cut way to so the same calculation. Let say I have following list object: List - vector('list', length = 3) set.seed(1) List[[1]] - rnorm(5) List[[2]] - rnorm(2) List[[3]] - rnorm(7) List [[1]] [1] -0.6264538 0.1836433 -0.8356286 1.5952808 0.3295078 [[2]] [1] -0.8204684 0.4874291 [[3]] [1] 0.7383247 0.5757814 -0.3053884 1.5117812 0.3898432 -0.6212406 -2.2146999 Vector - 3:5 Vector [1] 3 4 5 Now, what I want to do is, add List with Vector, element-by-element. Means I wanted to do: List[[1]] + Vector[1] [1] 2.373546 3.183643 2.164371 4.595281 3.329508 List[[2]] + Vector[2] [1] 3.179532 4.487429 List[[3]] + Vector[3] [1] 5.738325 5.575781 4.694612 6.511781 5.389843 4.378759 2.785300 Till now I have done this calculation with for-loop. Therefore it would be interesting if there is any elegant way to do the same. Thanks, __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] A Calculation on list object
Hi Ron, Seems like there might be a really elegant way, but I would use lapply(). For instance: lapply(seq_along(List), function(x) List[[x]] + Vector[x]) If you do this regularly and want something that reads more intuitively, consider defining an operator that does this. %+% is undefined (at least on my system), so something like: set.seed(1) List - list(rnorm(5), rnorm(2), rnorm(7)) Vector - 3:5 %+% - function(e1, e2) { if (identical(length(e1), length(e2))) lapply(seq_along(e1), function(i) e1[[i]] + e2[[i]]) else stop(length of e1 (, length(e1), ) must match length of e2 (, length(e2), ).) } List %+% Vector List %+% 11:13 This has the advantage of looking more like how you are thinking (add elements of the list to elements of the vector). Hope this helps, Josh On Mon, Jun 6, 2011 at 3:10 PM, Ron Michael ron_michae...@yahoo.com wrote: Hello, I am into some calculation on a list object, therefore requesting the peers if there is any short cut way to so the same calculation. Let say I have following list object: List - vector('list', length = 3) set.seed(1) List[[1]] - rnorm(5) List[[2]] - rnorm(2) List[[3]] - rnorm(7) List [[1]] [1] -0.6264538 0.1836433 -0.8356286 1.5952808 0.3295078 [[2]] [1] -0.8204684 0.4874291 [[3]] [1] 0.7383247 0.5757814 -0.3053884 1.5117812 0.3898432 -0.6212406 -2.2146999 Vector - 3:5 Vector [1] 3 4 5 Now, what I want to do is, add List with Vector, element-by-element. Means I wanted to do: List[[1]] + Vector[1] [1] 2.373546 3.183643 2.164371 4.595281 3.329508 List[[2]] + Vector[2] [1] 3.179532 4.487429 List[[3]] + Vector[3] [1] 5.738325 5.575781 4.694612 6.511781 5.389843 4.378759 2.785300 Till now I have done this calculation with for-loop. Therefore it would be interesting if there is any elegant way to do the same. Thanks, __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Joshua Wiley Ph.D. Student, Health Psychology University of California, Los Angeles http://www.joshuawiley.com/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] A Calculation on list object
On Mon, Jun 6, 2011 at 6:10 PM, Ron Michael ron_michae...@yahoo.com wrote: Hello, I am into some calculation on a list object, therefore requesting the peers if there is any short cut way to so the same calculation. Let say I have following list object: List - vector('list', length = 3) set.seed(1) List[[1]] - rnorm(5) List[[2]] - rnorm(2) List[[3]] - rnorm(7) List [[1]] [1] -0.6264538 0.1836433 -0.8356286 1.5952808 0.3295078 [[2]] [1] -0.8204684 0.4874291 [[3]] [1] 0.7383247 0.5757814 -0.3053884 1.5117812 0.3898432 -0.6212406 -2.2146999 Vector - 3:5 Vector [1] 3 4 5 Now, what I want to do is, add List with Vector, element-by-element. Means I wanted to do: List[[1]] + Vector[1] [1] 2.373546 3.183643 2.164371 4.595281 3.329508 List[[2]] + Vector[2] [1] 3.179532 4.487429 List[[3]] + Vector[3] [1] 5.738325 5.575781 4.694612 6.511781 5.389843 4.378759 2.785300 Till now I have done this calculation with for-loop. Therefore it would be interesting if there is any elegant way to do the same. Try this: mapply(+, List, Vector) -- Statistics Software Consulting GKX Group, GKX Associates Inc. tel: 1-877-GKX-GROUP email: ggrothendieck at gmail.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] draw text outside plot boundaries
Erik, To add to what epter said... I created this little function for clicking text anywhere on the plot (I probably stole the idea from a list serve or Dalgaard's book or someplace like that). Anyway it is helpful to me and may be of use to you too. Very basic but I use it a ton. You modify to suit your needs. #TEXT CLICK FUNCTION textClick-function(express,col=black,cex=NULL){ par(mar = rep(0, 4),xpd=NA) text(locator(1),express,col=col,cex=cex) } #EXAMPLE frame() par(mfrow=c(2,2)) with(mtcars,plot(mpg~cyl));with(mtcars,plot(mpg~cyl)) with(mtcars,plot(mpg~cyl));with(mtcars,plot(mpg~cyl)) textClick(expression(sum((bar(X)-X^2))),pink,.5) Cheers Tyler Date: Mon, 6 Jun 2011 15:09:29 -0700 From: ehl...@ucalgary.ca To: e...@q32.com CC: r-help@r-project.org Subject: Re: [R] draw text outside plot boundaries On 2011-06-06 08:44, Erik Aronesty wrote: i'd like to use the text() function to annotate some points, but the labels get cropped, if the point is on the right is there a way to prevent this, and tell the text() function to allow writing outside the boundaries of the current plot? Go to ?par and check out the 'xpd' parameter. Peter Ehlers i don't mind if it looks messy and steps on the margin a bit. - erik __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] extract data from a data frame field
Hi all, I am given the a data frame in which one of the columns has more information together- see column 4, peak_loc: chr startend peak_loc cluster_TC strand peak_TC 1 chr1 564620 564649 chr1:564644..564645,+ 94 + 10 2 chr1 565369 565404 chr1:565371..565372,+217 + 8 3 chr1 565463 565541 chr1:565480..565481,+ 1214 + 15 4 chr1 565653 565697 chr1:565662..565663,+ 1031 + 28 5 chr1 565861 565922 chr1:565883..565884,+316 + 12 6 chr1 566537 566573 chr1:566564..566565,+119 + 11 I am trying to find out if there's a way to extract the coordinates given in the 4th column and replace this column with two others that would have the start coord and the end coord. so instead of chr1:564644..564645,+ I would obtain; start_peak end_peak 564644 564645 Best, nanami [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] extract data from a data frame field
Here is a start; you can change the column names: x chr startend peak_loc cluster_TC strand peak_TC 1 chr1 564620 564649 chr1:564644..564645,+ 94 + 10 2 chr1 565369 565404 chr1:565371..565372,+217 + 8 3 chr1 565463 565541 chr1:565480..565481,+ 1214 + 15 4 chr1 565653 565697 chr1:565662..565663,+ 1031 + 28 5 chr1 565861 565922 chr1:565883..565884,+316 + 12 6 chr1 566537 566573 chr1:566564..566565,+119 + 11 y - sub(^.*:([[:digit:]]+)..([[:digit:]]+).*, \\1 \\2, x$peak_loc) y [1] 564644 564645 565371 565372 565480 565481 565662 565663 565883 565884 566564 566565 y - strsplit(y, ' ') y [[1]] [1] 564644 564645 [[2]] [1] 565371 565372 [[3]] [1] 565480 565481 [[4]] [1] 565662 565663 [[5]] [1] 565883 565884 [[6]] [1] 566564 566565 x.new - cbind(x, do.call(rbind, y)) x.new chr startend peak_loc cluster_TC strand peak_TC 1 2 1 chr1 564620 564649 chr1:564644..564645,+ 94 + 10 564644 564645 2 chr1 565369 565404 chr1:565371..565372,+217 + 8 565371 565372 3 chr1 565463 565541 chr1:565480..565481,+ 1214 + 15 565480 565481 4 chr1 565653 565697 chr1:565662..565663,+ 1031 + 28 565662 565663 5 chr1 565861 565922 chr1:565883..565884,+316 + 12 565883 565884 6 chr1 566537 566573 chr1:566564..566565,+119 + 11 566564 566565 On Mon, Jun 6, 2011 at 8:22 PM, ads pit deconstructed.morn...@gmail.com wrote: Hi all, I am given the a data frame in which one of the columns has more information together- see column 4, peak_loc: chr start end peak_loc cluster_TC strand peak_TC 1 chr1 564620 564649 chr1:564644..564645,+ 94 + 10 2 chr1 565369 565404 chr1:565371..565372,+ 217 + 8 3 chr1 565463 565541 chr1:565480..565481,+ 1214 + 15 4 chr1 565653 565697 chr1:565662..565663,+ 1031 + 28 5 chr1 565861 565922 chr1:565883..565884,+ 316 + 12 6 chr1 566537 566573 chr1:566564..566565,+ 119 + 11 I am trying to find out if there's a way to extract the coordinates given in the 4th column and replace this column with two others that would have the start coord and the end coord. so instead of chr1:564644..564645,+ I would obtain; start_peak end_peak 564644 564645 Best, nanami [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Jim Holtman Data Munger Guru What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Custom Sort on a Table object
Greetings - I've got the following table (the result of a two-way table operation): f m 0 to 5 11.328000 6.900901 15 to 24 6.100570 5.190058 25 to 34 9.428707 6.567280 35 to 4410.462158 7.513270 45 to 54 7.621988 5.692905 5 to 14 6.502741 6.119663 55 to 64 5.884737 4.319905 65 to 74 5.075606 4.267810 75 to 84 4.702020 3.602362 85 and over 4.75 3.877551 Which I'd like to sort so that the column of rownames (which represent age bands) and their corresponding f and m values appear in logical order. I've tried a bunch of things; merging with a separate df bearing Age Bands paired with a sequence number, stripping out row vectors and rbind-ing a new df, etc., all to no avail. It seems to be very difficult to spin a table object into a data frame without being stuck with the tables rownames! I haven't yet tried writing to an external file and then reading it back (so as to get R to forget that it's a Table object), and then merging on the group bands to pull in a sequence vector upon which to do an order(). Seems like it should be easier. Many thanks, Galen [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Custom Sort on a Table object
Here is a one way. tab fm 0 to 5 11.328000 6.900901 15 to 24 6.100570 5.190058 25 to 34 9.428707 6.567280 35 to 4410.462158 7.513270 45 to 54 7.621988 5.692905 5 to 14 6.502741 6.119663 55 to 64 5.884737 4.319905 65 to 74 5.075606 4.267810 75 to 84 4.702020 3.602362 85 and_over 4.75 3.877551 lowAge - as.numeric(sapply(strsplit(rownames(tab), ), [, 1)) (tab - tab[order(lowAge), ]) fm 0 to 5 11.328000 6.900901 5 to 14 6.502741 6.119663 15 to 24 6.100570 5.190058 25 to 34 9.428707 6.567280 35 to 4410.462158 7.513270 45 to 54 7.621988 5.692905 55 to 64 5.884737 4.319905 65 to 74 5.075606 4.267810 75 to 84 4.702020 3.602362 85 and over 4.75 3.877551 -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Galen Moore Sent: Tuesday, 7 June 2011 1:23 PM To: r-help@r-project.org Subject: [R] Custom Sort on a Table object Greetings - I've got the following table (the result of a two-way table operation): f m 0 to 5 11.328000 6.900901 15 to 24 6.100570 5.190058 25 to 34 9.428707 6.567280 35 to 4410.462158 7.513270 45 to 54 7.621988 5.692905 5 to 14 6.502741 6.119663 55 to 64 5.884737 4.319905 65 to 74 5.075606 4.267810 75 to 84 4.702020 3.602362 85 and over 4.75 3.877551 Which I'd like to sort so that the column of rownames (which represent age bands) and their corresponding f and m values appear in logical order. I've tried a bunch of things; merging with a separate df bearing Age Bands paired with a sequence number, stripping out row vectors and rbind-ing a new df, etc., all to no avail. It seems to be very difficult to spin a table object into a data frame without being stuck with the tables rownames! I haven't yet tried writing to an external file and then reading it back (so as to get R to forget that it's a Table object), and then merging on the group bands to pull in a sequence vector upon which to do an order(). Seems like it should be easier. Many thanks, Galen [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Line Graphs
Hello, I want to plot 6 line graphs. I have 10 points 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9 and 1.0. At each point say 0.1, I have 6 variables A, B, C, D, E and F. The variables all have values between 0 and 1 (and including 0 and 1). I also want to label the x axis from 0.1 to 1.0 and the y axis from 0.1 to 1.0. My goal is to plot a line graph representing the mean of the variables at each level. SO for 0.1 on the xaxis, we should expect 6 values for y. This is what I have so far. The plot omits 1.0 and the abline function does not make the line y = x on the plot This is what I have so far: # Calculate range from 0 to max value of cars and trucks g_range - range(0, 1) # Graph autos using y axis that ranges from 0 to max # value in cars or trucks vector. Turn off axes and # annotations (axis labels) so we can specify them ourself plot(B, type=o, pch = 0, lty=1,col=blue, ylim=g_range, axes=FALSE, ann=FALSE) # Make x axis using the values of pi_0 labels # axis(1, at=1:10, lab=c(0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.0)) # Make y axis with horizontal labels that display ticks at del = seq(0.1,1, 0.1) axis(2, at=del, lab=c(0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.0)) # Create box around plot box() # Graph trucks with red dashed line and square points lines(A, type=o, pch=2, lty=1, col=red) lines(C, type=o, pch=3, lty=1, col=green) lines(D, type=o, pch=4, lty=1, col=orange) lines(E, type=o, pch=6, lty=1, col=brown) lines(F, type=o, pch=8, lty=1, col=yellow) abline(0, 1, col = black) # Create a title with a red, bold/italic font #title(main=Methods, col.main=red, font.main=4) # Label the x and y axes title(xlab=expression(paste(lambda[0]))) title(ylab= expression(paste(Estimate of , lambda[0]))) # Create a legend at (1, g_range[2]) that is slightly smaller # (cex) and uses the same line colors and points used by # the actual plots legend(1, g_range[2], c(B,A, C, D, E, F, Actual), cex=0.6, col=c(blue,red, green, orange, brown,yellow, black), pch=c(0,2,3,4,6,8,9), lty=1) -- Thanks, Jim. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] error with geomap in googleVis
Hi All, I am unable to get the plot geomap in googleVis package. data is as follows head(index.ret) countryytd 1 Argentina -10.18 2 Australia -3.42 3 Austria -2.70 4 Belgium 1.94 5Brazil -7.16 6Canada 0.56 map1 = gvisGeoMap(index.ret,locationvar = 'country', numvar = 'ytd') plot(map1) But it just displays a blank page, showing an error symbol at the right bottom corner. I tried demo(googleVis), it also had a similar problem. The demo showed all other plots/maps except for those geomaps. Could any one please hint me what/where could be the problem? Many thanks for the idea and support. Regards, SNV Krishna [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] extract data features from subsets
I have a large dataset similar to this: ID timeresult A 1 5 A 2 2 A 3 1 A 4 1 A 5 1 A 6 2 A 7 3 A 8 4 B 1 3 B 2 2 B 3 4 B 4 6 B 5 8 I need to extract a number of features for each individual in it (identified by ID). These are: * The lowest result (the nadir) * The time of the nadir - but if the nadir level is present at 1 time point, I need the minimum and maximum time of nadir * For the time period from maximum time of nadir to the last result, I need the coefficient from a lm(result~time) The result would be a table looking like: ID NadirLevel NadirFirstTime NadirLastTime Slope A 1 3 5 1 B 2 2 2 2 I can manage to extract all the required elements in a very cumbersome loop, but I am sure an elegant method using apply() or the like could be devised but I cant presently understand the necessary syntax. An suggestions appreciated. Thanks Scott _ Dr. Scott Williams Peter MacCallum Cancer Centre Melbourne, Australia ph +61 3 9656 fax +61 3 9656 1424 scott.willi...@petermac.org This email (including any attachments or links) may contain confidential and/or legally privileged information and is intended only to be read or used by the addressee. If you are not the intended addressee, any use, distribution, disclosure or copying of this email is strictly prohibited. Confidentiality and legal privilege attached to this email (including any attachments) are not waived or lost by reason of its mistaken delivery to you. If you have received this email in error, please delete it and notify us immediately by telephone or email. Peter MacCallum Cancer Centre provides no guarantee that this transmission is free of virus or that it has not been intercepted or altered and will not be liable for any delay in its receipt. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.