Re: [R] Can't there be a cd command?
Issac Trotts [EMAIL PROTECTED] writes: I'm on Linux, so it didn't occur to me to look in the Windows FAQ until it came up in the Google search. Why should Windows users be the only ones who learn how to change directories? Well, on Linux et al., the normal mode of operation is to start R from the commmand line in the directory where you want to be. On Windows you start from a desktop icon or a menu entry, which implies that you're always in the same directory, and usually the wrong one. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] microsoft proxy R project
Thanks a lot, you are nice. Laurent. -Original Message- From: Nolwenn LeMeur [mailto:[EMAIL PROTECTED] Sent: lundi 8 mai 2006 08:16 To: Laurent Valdes Cc: r-help@stat.math.ethz.ch; Remy Drouilhet; Valdes, Laurent; [EMAIL PROTECTED] Subject: Re: [R] microsoft proxy R project Hi Laurent, You should look the R windows FAQ web page: (a) Use the alternative |internet2.dll| by starting R with the flag --internet2 (see How do I install R for Windows? http://cran.r-project.org/bin/windows/base/rw-FAQ.html#How-do-I-install-R-for-Windows_003f) which uses the Internet Explorer internals (and so needs Internet Explorer 4 or later installed). Note that this does not work with proxies that need authentication. (b) A proxy needs to be set up: see |?download.file|. Here are two versions of an example (a real one, but from a machine that is only available locally) of a command-line in a short cut: path_to_R\bin\RGui.exe http_proxy=http://user:[EMAIL PROTECTED]:80/ path_to_R\bin\RGui.exe http_proxy=http://gannet/ http_proxy_user=ask The second version will prompt the user for the proxy username and password when HTTP downloads are first used. Nolwenn Laurent Valdes wrote: hi, as I'm using R at Dell (where I am working), I manage to install packages from web, but there is a web proxy, which is pretty boring. R asks me to enter an active directory login/password, that would normally auth me through the web proxy, but it doesn't work. I'm looking forward for your help. Sorry if I used bad english. Any ideas ? Laurent -- «Un peu d'internationalisme éloigne de la patrie; beaucoup y ramène» Jean Jaurès@http://www.le-valdo.com [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] warning messages ?
hello I received these warning messages while I was building logistic regression model. Could you tell me why this warnings arises ? regards Warning messages: 1: algorithm did not converge in: glm.fit(x = X, y = Y, weights = weights, start = start, etastart = etastart, 2: fitted probabilities numerically 0 or 1 occurred in: glm.fit(x = X, y = Y, weights = weights, start = start, etastart = etastart, -- Ahmet Temiz Jeoloji Müh. Afet İşleri Genel Müdürlüğü Deprem Araştırma Dairesi Tel: (312) 287 89 51 veya (312) 287 26 80/1547 Faks: (312) 287 89 51 E. Posta: [EMAIL PROTECTED] www.deprem.gov.tr Ahmet Temiz Geological Eng. General Directorate of Disaster Affairs Earthquake Research Department Phone: +90 (312) 287 89 51 or (312) 287 26 80/1547 Fax: +90 (312) 287 89 51 E. Mail: [EMAIL PROTECTED] www.deprem.gov.tr -- This message has been scanned for viruses and\ dangerous con...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On Wed, 10 May 2006, Peter Dalgaard wrote: Issac Trotts [EMAIL PROTECTED] writes: I'm on Linux, so it didn't occur to me to look in the Windows FAQ until it came up in the Google search. Why should Windows users be the only ones who learn how to change directories? Well, on Linux et al., the normal mode of operation is to start R from the commmand line in the directory where you want to be. On Windows you start from a desktop icon or a menu entry, which implies that you're always in the same directory, and usually the wrong one. Exactly. I don't think I have ever used setwd() on Linux. Also, I have never seen this asked before for a Unix-alike, so it seems not to be a FAQ. There is a common tendency for users who run into a problem to think everyone does too, and it isn't necessarily so. Frequently asked questions do make it to the FAQs: it is a defence mechanism for the volunteers supporting R. help.search(directory) gets you there. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] new package error message
On Tue, 9 May 2006, Jason Liao wrote: My coauthor made a new R package GeneLogit (100% R code) which installs ok on R 2.3 on Windows. But when I type library(GeneLogit) it gave the error message Error in library(GeneLogit) : 'GeneLogit' is not a valid package -- installed 2.0.0? It runs on R 1.9.0 just fine. It seems that others have encountered same problem but no solution is found by googling In every case I have ever seen mentioned, the solution was to install the package properly by R CMD INSTALL and not by a manual kludge. The message is given if the installation lacks the metadata in the Meta subdirectory which INSTALL creates. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Full or non constrained/reparametrized model.matrix
Gabor thanks! Last suggestion works like charm! Gabor Grothendieck wrote: On 5/9/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: On 5/9/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! Thank you very much for the response. Please read within the lines Gabor Grothendieck wrote: On 5/9/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! I have parameter estimates for a generalized linear model and would like to produce fitted values i.e. fitted(). This can be easily done in R, but my problem lies in fact that I have a vector of parameters from some other software, that is is not constrained i.e. I have the following estimates for model with one factor with 4 levels beta = c(intercept group1 group2 group3 group4) where group1:4 are estimated deviations from intercept i.e. sum to zero contraint, but all parameter estimates are there! How can I create a model matrix that will not contain any constraints since I would like to compute model.matrix(some_formula) %*% beta I.e. I would like to have model.matrix of the form 1 1 0 0 0 1 0 1 0 0 1 0 0 1 0 1 0 0 0 1 and not of the following form with contr.treatment or any other contraints 1 0 0 0 1 1 0 0 1 0 1 0 1 0 0 1 I could remove group4 from beta and use sum to zero contraint for contrast in fomula, but I would like to overcome this, as my model can be richer in number or parameters. The following example, will show what I would like to do: ## --- Setup --- groupN - 4 NPerGroup - 5 min - 1 max - 5 g - runif(n = groupN, min = min, max = max) ## --- Simulate --- group - factor(rep(paste(G, 1:groupN, sep = ), each = NPerGroup)) y - vector(mode = numeric, length = groupN * NPerGroup) j - 1 for (i in 1:groupN) { y[j:(i * NPerGroup)] - rpois(n = NPerGroup, lambda = g[i]) j - (i * NPerGroup) + 1 } ## --- GLM --- contrasts(group) - contr.sum(groupN) fit - glm(y ~ group, family = poisson) coef(fit) ## Now this goes nicely model.matrix(y ~ group) %*% coef(fit) ## But pretend I have the following vector of estimated parameters beta - c(coef(fit), 0 - sum(coef(fit)[-1])) names(beta) - c(names(beta)[1:4], group4) ## I can not apply this as model matrix does not conform to beta model.matrix(y ~ group) %*% beta Try this: model.matrix(y ~ group-1) %*% beta[-1] + beta[1] This is a nice hack here, but does not work in general. Say I have another factor sex - factor(rep(paste(S, 1:2, sep = ), times=10)) model.matrix(y ~ group + sex - 1) produces a matrix of 5 columns and not 6 as I want to. Try this: cbind(1, model.matrix(~group-1), model.matrix(~sex-1)) In thinking about this a bit more, try this: attr(group, contrasts) - diag(nlevels(group)) attr(sex, contrasts) - diag(nlevels(sex)) model.matrix(~ group + sex) -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Unix X-WIndows RGUI
On Tue, 9 May 2006, Alex Restrepo wrote: Could someone tell me if there is an X-Windows version of the RGUI ? There is not. An one point there was a Motif version (graphapp, the toolkit used for RGui, used to have a Motif version), but it was incomplete (and unmaintainable). There are several other consoles for R which work under X Windows. Package gnomeGUI is one that used to be bundled in R, and for others see http://www.sciviews.org/_rgui/. [JGR is the one that is most frequently mentioned (but you will need working Java, which precludes my current x86_64 Linux platform).] -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] How to use function to change density in sm.density?
I have trawl data (fish) for different areas. In each catch length of the different fish is registerd. However the real catch has to be calibrated because the trawl is selective. For fish=15cm the adjusted catch is (old.frequency*35/19) for fish15cm and 62cm the adjusted catch is (old.frequency*35/( 5.91*frequency^0.43) and for fish=62cm the catch need no adjustment. Basically what I need is to know how to write an expression i R that says something like if (fishlength=15) new.dens-old.density*35/19 if (fishlength15fishlength62) update the new.dens with old.dens*35/( 5.91*density^0.43) else old.density=old.density (this should also be updated in new.dens) After these steps I wold like to draw the new adjusted densityline. Any suggestions are welcome. Best regards T M Lunde [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/9/06, Prof Brian Ripley [EMAIL PROTECTED] wrote: On Wed, 10 May 2006, Peter Dalgaard wrote: Issac Trotts [EMAIL PROTECTED] writes: I'm on Linux, so it didn't occur to me to look in the Windows FAQ until it came up in the Google search. Why should Windows users be the only ones who learn how to change directories? Well, on Linux et al., the normal mode of operation is to start R from the commmand line in the directory where you want to be. On Windows you start from a desktop icon or a menu entry, which implies that you're always in the same directory, and usually the wrong one. Exactly. I don't think I have ever used setwd() on Linux. Also, I have never seen this asked before for a Unix-alike, so it seems not to be a FAQ. There is a common tendency for users who run into a problem to think everyone does too, and it isn't necessarily so. Frequently asked questions do make it to the FAQs: it is a defence mechanism for the volunteers supporting R. help.search(directory) gets you there. OK, good to know. Thanks for helping. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Until the key pressed: FOR-LOOP-Clustering
Hi All, These are the code that i used to plot the kmeans clustering. DataSetS01022-rbind(matrix(rnorm(50),ncol=2), + matrix(rnorm(50),ncol=2)) colnames(DataSetS01022) -c(timeslot,em) (cl -kmeans(DataSetS01022,2)) plot (DataSetS01022, col=cl$cluster) points(cl$centers, col = 1:2, pch = 8, cex=2) Now, i would like to do *the same process for 5 times*, so that i can view the clusters in data have been moving and the center changes with various results. (Animation-like) My main concern here, is how to put this in a *loop form* or any other methods are available? I tried with:* clarax$clusinfo I thought of doing it to activate Device (1) (2)* ** *Really need anyones help!* ** *JJ* *---* -- Lecturer J. Joshua Thomas KDU College Penang Campus Research Student, University Sains Malaysia [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] remove Punctuation characters
Thanks a lot!! Filipe Almeida Marc Schwartz (via MN) wrote: On Tue, 2006-05-09 at 16:50 +0100, Filipe Almeida wrote: Hi, I want to remove all punctuation characters in a string. I was trying it use a regular expressions but it doesn't work. Here is a sample os what i want: str - 'ABD - remove de punct, and dot characters.' str - gsub('[:punct:]','',str) str 'ABD remove de punct and dot characters is there any function that do this kind of thing? Thanks to all. Filipe Almeida You almost have it. Just need to double the brackets: str [1] ABD - remove de punct, and dot characters. gsub([[:punct:]], , str) [1] ABD remove de punct and dot characters Note the following in ?regex: For example, [[:alnum:]] means [0-9A-Za-z], except the latter depends upon the locale and the character encoding, whereas the former is independent of locale and character set. (Note that the brackets in these class names are part of the symbolic names, and must be included in addition to the brackets delimiting the bracket list.) Most metacharacters lose their special meaning inside lists. To include a literal ], place it first in the list. Similarly, to include a literal ^, place it anywhere but first. Finally, to include a literal -, place it first or last. (Only these and \ remain special inside character classes.) HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
Issac Trotts wrote: On 5/9/06, Prof Brian Ripley [EMAIL PROTECTED] wrote: On Wed, 10 May 2006, Peter Dalgaard wrote: Issac Trotts [EMAIL PROTECTED] writes: I'm on Linux, so it didn't occur to me to look in the Windows FAQ until it came up in the Google search. Why should Windows users be the only ones who learn how to change directories? Well, on Linux et al., the normal mode of operation is to start R from the commmand line in the directory where you want to be. On Windows you start from a desktop icon or a menu entry, which implies that you're always in the same directory, and usually the wrong one. Exactly. I don't think I have ever used setwd() on Linux. Also, I have never seen this asked before for a Unix-alike, so it seems not to be a FAQ. There is a common tendency for users who run into a problem to think everyone does too, and it isn't necessarily so. Frequently asked questions do make it to the FAQs: it is a defence mechanism for the volunteers supporting R. help.search(directory) gets you there. OK, good to know. Thanks for helping. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html probably this has used up sufficient disk space in the mail archive already, but anyway a bit of support to the initial request: one of the very first things I did, when I started using R, was adding cd - setwd pwd - getwd to my .Rprofile. and when I encouraged colleagues to try R out, the 'top three' questions where how can I get out? I tried ^C, exit(), quit(), but nothing works how can I change the directory? `cd' does'nt work where am I? `pwd' does'nt work (`pwd' is something like a 'local speciality', I presume...) this is of course a marginal obstacle for new users, but avoidable all the same. (and maybe most people (like myself) in the first place don't think it important enough to post it -- maybe it _would_ be in the FAQ otherwise...) so, _if_ 'cd' would be recognized in future releases, it would'nt do any harm, would it? joerg __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] subselect in a matrix
hello I confused one point in matrix constructed with cbind I want to subselect 6. column 1000 what should the syntax be ? regards -- Ahmet Temiz Jeoloji Müh. Afet İşleri Genel Müdürlüğü Deprem Araştırma Dairesi Tel: (312) 287 89 51 veya (312) 287 26 80/1547 Faks: (312) 287 89 51 E. Posta: [EMAIL PROTECTED] www.deprem.gov.tr Ahmet Temiz Geological Eng. General Directorate of Disaster Affairs Earthquake Research Department Phone: +90 (312) 287 89 51 or (312) 287 26 80/1547 Fax: +90 (312) 287 89 51 E. Mail: [EMAIL PROTECTED] www.deprem.gov.tr -- This message has been scanned for viruses and\ dangerous con...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] subselect in a matrix
orkun [EMAIL PROTECTED] writes: hello I confused one point in matrix constructed with cbind I want to subselect 6. column 1000 what should the syntax be ? m[m[, 6] 1000, ] if I understand your question correctly. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] draw a graphic using segments to represent data
Wang, Xuetao wrote: Hello, I'd like to draw a graphic using different color line segments to represent microarray data. It simply likes this: Data=cbind(c(1, 2, -1, 3, -2.2), c(1, 3, -1, -1.8, 4)) The first col No represents y'axis position (negative down and positive up). The second col No represents a length of segment (negative segment should be on left of y, and positive on right). Also the segment color on I, II, III, IV should be different. Something like myplot - function(x, y, segcol=1:4, col, ...){ plot(x, y, type=n, ...) abline(h=0, v=0, col=grey, ...) if(missing(col)) col - segcol[1 + (x 0) + 2 * (y 0)] segments(x, y, 0, y, col = col, ...) } myplot(Data[,1], Data[,2]) springs to mind. Thank you for your time and help. Xueato [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] transposing a big data file
yohannes alazar wrote: I HAVE A VERY BIG DATA OF 67 COLMS AND 25000 ROWS AND WOULD LIKE TO TRANSPOSE IT THE R HELP WAS NOT ENOUGH INFORMATION BECOUSE I AM NOT A PROGRAMMER AND FIRST TIME R USER. SO CAN YOU GIVE SOME HINTS OF CODING, You might want to fix your Caps Lock key ... AA TT GG GG CC AA TT GG GG CC AA TT GG GG CC AA TT GG GG CC AA TT GG GG CC Looks like 1x25 TO AA AA AA AA AA TT TT TT TT TT GG GG GG GG GG GG GG GG GG GG CC CC CC CC CC 1x25 ... Do you really want to transpose or to rearrange in some other manner? Transposing is done by t() as help.search(transpose) turns out Uwe Ligges [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R and Java
hello, I want to call R from java. I know that i must install SJava .so ,what tools i need and how i install hwo help me ? [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R and Java
You keep on asking questions on this topic lacking basic information. Please consult the posting guide: 1) Do not send HTML mail. 2) Do tell us your OS and platform. 3) Do ask the maintainer about contributed packages. There are alternatives to SJava, for example Rserve and JRI at http://www.rosuda.org/software/ (and see the R FAQ). On Wed, 10 May 2006, miniar mansouri wrote: hello, I want to call R from java. I know that i must install SJava .so ,what tools i need and how i install hwo help me ? [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] transposing a big data file
I think that's a 5 X 5 matrix. However, it is very difficult to read words in Caps. On 5/10/06, Uwe Ligges [EMAIL PROTECTED] wrote: yohannes alazar wrote: I HAVE A VERY BIG DATA OF 67 COLMS AND 25000 ROWS AND WOULD LIKE TO TRANSPOSE IT THE R HELP WAS NOT ENOUGH INFORMATION BECOUSE I AM NOT A PROGRAMMER AND FIRST TIME R USER. SO CAN YOU GIVE SOME HINTS OF CODING, You might want to fix your Caps Lock key ... AA TT GG GG CC AA TT GG GG CC AA TT GG GG CC AA TT GG GG CC AA TT GG GG CC Looks like 1x25 TO AA AA AA AA AA TT TT TT TT TT GG GG GG GG GG GG GG GG GG GG CC CC CC CC CC 1x25 ... Do you really want to transpose or to rearrange in some other manner? Transposing is done by t() as help.search(transpose) turns out Uwe Ligges [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] draw a graphic using segments to represent data
Hi, I am not sure whether I completely understood your question. Anyway, here it's a simple code that, hopefully, overcome your problem. Data=cbind(c(1, 2, -1, 3, -2.2), c(1, 3, -1, -1.8, 4)) plot(x=rep(0,nrow(Data)), Data[,1], axes=F, xlim=c(-max(Data[,2]), max(Data[,2])), xlab=, col=1:nrow(Data), cex=2) axis(2) for(i in 1:nrow(Data)){ segments(x0=0, y0=Data[i,1], x1=Data[i,1] + Data[i,2], y1=Data[i,1], col=i) } Best, Carlo Giovani Camarda === Camarda Carlo Giovanni PhD-Student Max Planck Institute for Demographic Research Konrad-Zuse-Strasse 1 18057 Rostock, Germany Tel: +49 (0)381 2081 172 Fax: +49 (0)381 2081 472 [EMAIL PROTECTED] === Date: Tue, 9 May 2006 15:10:37 -0400 From: Wang, Xuetao [EMAIL PROTECTED] Subject: [R] draw a graphic using segments to represent data To: r-help@stat.math.ethz.ch Message-ID: [EMAIL PROTECTED] Content-Type: text/plain Hello, I'd like to draw a graphic using different color line segments to represent microarray data. It simply likes this: Data=cbind(c(1, 2, -1, 3, -2.2), c(1, 3, -1, -1.8, 4)) The first col No represents y'axis position (negative down and positive up). The second col No represents a length of segment (negative segment should be on left of y, and positive on right). Also the segment color on I, II, III, IV should be different. Thank you for your time and help. Xueato -- This mail has been sent through the MPI for Demographic Rese...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] identify high dimension data point
Hi there, I am having some trouble with the Identify() function, it looks that the identify() function only works on the plot of 2-dimension data set. In high-dimension case , if I use pairs() + identify() in hope to observe all the components of the data point I selected, I get error message. Is there anyway to get around this ? Plus, I also need it to actually compare all the components of the picked data point with mydataset to choose the closest one. Thanks million in advance. best tong __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] kernel smoothing of disease rates at locations
I want to do kernel smoothing on the standardised mortality rates at specified location(x-y coordinates) and be able to produce contour map. I've tried to use the package called fields in R (function smooth.2d) but the problem is I get negative results after smoothing and I do not understand why? Your contributions will be highly appreciated or if there is any package that can do this I will be happy to know about it. Thanks in advance for your comments. Oarabile Below is how my data in R looks like XcoordYcoord smr 1 383894.70 653202.8 1.5816327 2 339053.00 631770.9 1.0003473 3 369164.50 609973.2 1.2943872 4 313477.70 640382.8 0.9443508 5 290748.80 695088.8 1.1662125 6 287547.50 679829.6 0.9433962 7 259031.90 704478.3 1.0184905 8 329107.60 585753.3 1.0445860 9 292170.20 578978.2 1.1287758 10 268330.90 564281.7 0.9805996 11 227395.50 561524.6 1.1947431 12 310130.70 689395.0 1.0991957 13 325756.00 696565.6 1.1734694 14 339427.20 713217.2 0.8724100 15 391752.10 805661.5 1.2483221 16 385935.90 851044.5 1.0558140 17 372471.20 826373.7 1.4220183 18 354901.70 791107.2 0.9175084 19 324944.10 842040.8 1.0447080 20 275672.50 802570.4 0.9899329 21 320750.00 954769.7 1.0866834 22 243489.20 832748.8 1.0995658 23 166559.70 786365.3 1.0401267 24 288160.80 846054.4 1.0299003 25 220347.00 869181.0 0.9529435 26 175863.63 822593.9 0.9902252 27 246114.40 938419.3 0.8909370 28 359070.60 669738.3 0.9883721 29 320614.70 670434.2 1.1288534 30 331414.80 659972.3 0.9070866 31 299336.90 666122.9 0.7969349 32 156965.25 708047.7 1.1410460 33 253867.30 674916.5 0.9256710 34 249188.40 674272.8 1.0869565 35 272463.00 679495.3 0.8576698 36 252643.80 611378.4 0.8615611 37 212637.35 643725.0 0.8596604 38 234751.95 691803.4 0.9706546 39 264856.20 644684.8 0.8468395 40 253949.30 653042.5 0.8106169 41 260412.35 663552.2 0.9253686 42 270807.60 657454.0 0.9002826 43 228477.30 670443.2 1.0782502 44 248215.30 641897.3 0.9240407 45 229797.50 596059.0 1.1579197 46 291725.00 635771.8 0.8908766 47 278012.80 667982.0 1.024 48 279490.80 659761.6 0.9828301 49 242791.70 662651.8 1.0230584 50 263217.20 675295.2 0.9759817 51 341138.80 754471.3 0.9505541 52 334963.00 735075.4 0.9264498 53 286139.70 733611.6 0.9052453 54 354015.60 1025696.8 0.8905506 55 455775.60 1201636.0 1.0862620 56 88429.32 854994.4 0.9577015 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] using crr in cmprsk
Hi, I need to fit model using crr, however my covariate is categorical with 3 levels. I use crr(time,status,agesplit,failcode=1,cencode=0) where agesplit is defined as 20,21-29,30 years, so it takes 0, 1 or 2 for each patient. I hoped to get estimated coefficients for the levels 1 and 2 w.r.t level 0 as in coxph. But, I didn't. Could someone please help me to use crr in this situation? Thanks a lot. Ruwanthi __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] kernel smoothing of disease rates at locations
It is certainly possible to get negative values when smoothing, especially on the edges. But we cannot reproduce your problem because you did not provide any code. -roger Oarabile Molaodi wrote: I want to do kernel smoothing on the standardised mortality rates at specified location(x-y coordinates) and be able to produce contour map. I've tried to use the package called fields in R (function smooth.2d) but the problem is I get negative results after smoothing and I do not understand why? Your contributions will be highly appreciated or if there is any package that can do this I will be happy to know about it. Thanks in advance for your comments. Oarabile Below is how my data in R looks like XcoordYcoord smr 1 383894.70 653202.8 1.5816327 2 339053.00 631770.9 1.0003473 3 369164.50 609973.2 1.2943872 4 313477.70 640382.8 0.9443508 5 290748.80 695088.8 1.1662125 6 287547.50 679829.6 0.9433962 7 259031.90 704478.3 1.0184905 8 329107.60 585753.3 1.0445860 9 292170.20 578978.2 1.1287758 10 268330.90 564281.7 0.9805996 11 227395.50 561524.6 1.1947431 12 310130.70 689395.0 1.0991957 13 325756.00 696565.6 1.1734694 14 339427.20 713217.2 0.8724100 15 391752.10 805661.5 1.2483221 16 385935.90 851044.5 1.0558140 17 372471.20 826373.7 1.4220183 18 354901.70 791107.2 0.9175084 19 324944.10 842040.8 1.0447080 20 275672.50 802570.4 0.9899329 21 320750.00 954769.7 1.0866834 22 243489.20 832748.8 1.0995658 23 166559.70 786365.3 1.0401267 24 288160.80 846054.4 1.0299003 25 220347.00 869181.0 0.9529435 26 175863.63 822593.9 0.9902252 27 246114.40 938419.3 0.8909370 28 359070.60 669738.3 0.9883721 29 320614.70 670434.2 1.1288534 30 331414.80 659972.3 0.9070866 31 299336.90 666122.9 0.7969349 32 156965.25 708047.7 1.1410460 33 253867.30 674916.5 0.9256710 34 249188.40 674272.8 1.0869565 35 272463.00 679495.3 0.8576698 36 252643.80 611378.4 0.8615611 37 212637.35 643725.0 0.8596604 38 234751.95 691803.4 0.9706546 39 264856.20 644684.8 0.8468395 40 253949.30 653042.5 0.8106169 41 260412.35 663552.2 0.9253686 42 270807.60 657454.0 0.9002826 43 228477.30 670443.2 1.0782502 44 248215.30 641897.3 0.9240407 45 229797.50 596059.0 1.1579197 46 291725.00 635771.8 0.8908766 47 278012.80 667982.0 1.024 48 279490.80 659761.6 0.9828301 49 242791.70 662651.8 1.0230584 50 263217.20 675295.2 0.9759817 51 341138.80 754471.3 0.9505541 52 334963.00 735075.4 0.9264498 53 286139.70 733611.6 0.9052453 54 354015.60 1025696.8 0.8905506 55 455775.60 1201636.0 1.0862620 56 88429.32 854994.4 0.9577015 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Increase the number of plotting characters in dotplot (lattice)?
Dear list, I am using the dotplot function of lattice and like it very much. However, in some of the plots, I have more than 7 categories that I would like to represent by different plotting characters in the graph, which is a problem. For the categories after the 7th, the plotting characters are recycled, creating a confisuing graph indeed. So, my question is: how do I increase the number of unique plotting characters? /Fredrik __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Increase the number of plotting characters in dotplot (lattice)?
library(lattice) show.settings() trellis.par.set(superpose.symbol = list(pch = 1:16), dot.symbol = list(pch = 1:16)) show.settings() ?trellis.par.get Fredrik Karlsson wrote: Dear list, I am using the dotplot function of lattice and like it very much. However, in some of the plots, I have more than 7 categories that I would like to represent by different plotting characters in the graph, which is a problem. For the categories after the 7th, the plotting characters are recycled, creating a confisuing graph indeed. So, my question is: how do I increase the number of unique plotting characters? /Fredrik __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 512-0171 (M, W, F) fax: (917) 438-0894 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] kernel smoothing of disease rates at locations
On Wed, 10 May 2006, Roger D. Peng wrote: It is certainly possible to get negative values when smoothing, especially on the edges. But we cannot reproduce your problem because you did not provide any code. But kernel smoothing always reduces the range of the data, so kernel smoothing of non-negative data cannot give a negative estimate. Roger's comments apply to other smoothers, e.g. local polynomials. We do need to see reproducible code (and also the posting guide advises contacting the package maintainer about problems with contributed packages). -roger Oarabile Molaodi wrote: I want to do kernel smoothing on the standardised mortality rates at specified location(x-y coordinates) and be able to produce contour map. I've tried to use the package called fields in R (function smooth.2d) but the problem is I get negative results after smoothing and I do not understand why? Your contributions will be highly appreciated or if there is any package that can do this I will be happy to know about it. Thanks in advance for your comments. Oarabile Below is how my data in R looks like XcoordYcoord smr 1 383894.70 653202.8 1.5816327 2 339053.00 631770.9 1.0003473 3 369164.50 609973.2 1.2943872 4 313477.70 640382.8 0.9443508 5 290748.80 695088.8 1.1662125 6 287547.50 679829.6 0.9433962 7 259031.90 704478.3 1.0184905 8 329107.60 585753.3 1.0445860 9 292170.20 578978.2 1.1287758 10 268330.90 564281.7 0.9805996 11 227395.50 561524.6 1.1947431 12 310130.70 689395.0 1.0991957 13 325756.00 696565.6 1.1734694 14 339427.20 713217.2 0.8724100 15 391752.10 805661.5 1.2483221 16 385935.90 851044.5 1.0558140 17 372471.20 826373.7 1.4220183 18 354901.70 791107.2 0.9175084 19 324944.10 842040.8 1.0447080 20 275672.50 802570.4 0.9899329 21 320750.00 954769.7 1.0866834 22 243489.20 832748.8 1.0995658 23 166559.70 786365.3 1.0401267 24 288160.80 846054.4 1.0299003 25 220347.00 869181.0 0.9529435 26 175863.63 822593.9 0.9902252 27 246114.40 938419.3 0.8909370 28 359070.60 669738.3 0.9883721 29 320614.70 670434.2 1.1288534 30 331414.80 659972.3 0.9070866 31 299336.90 666122.9 0.7969349 32 156965.25 708047.7 1.1410460 33 253867.30 674916.5 0.9256710 34 249188.40 674272.8 1.0869565 35 272463.00 679495.3 0.8576698 36 252643.80 611378.4 0.8615611 37 212637.35 643725.0 0.8596604 38 234751.95 691803.4 0.9706546 39 264856.20 644684.8 0.8468395 40 253949.30 653042.5 0.8106169 41 260412.35 663552.2 0.9253686 42 270807.60 657454.0 0.9002826 43 228477.30 670443.2 1.0782502 44 248215.30 641897.3 0.9240407 45 229797.50 596059.0 1.1579197 46 291725.00 635771.8 0.8908766 47 278012.80 667982.0 1.024 48 279490.80 659761.6 0.9828301 49 242791.70 662651.8 1.0230584 50 263217.20 675295.2 0.9759817 51 341138.80 754471.3 0.9505541 52 334963.00 735075.4 0.9264498 53 286139.70 733611.6 0.9052453 54 354015.60 1025696.8 0.8905506 55 455775.60 1201636.0 1.0862620 56 88429.32 854994.4 0.9577015 -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] identify high dimension data point [Broadcast]
You might want to give Ggobi a try (www.ggobi.org). There's a rggobi package that lets you run Ggobi on data in R, or control Ggobi from within R. Andy From: Tong Wang Hi there, I am having some trouble with the Identify() function, it looks that the identify() function only works on the plot of 2-dimension data set. In high-dimension case , if I use pairs() + identify() in hope to observe all the components of the data point I selected, I get error message. Is there anyway to get around this ? Plus, I also need it to actually compare all the components of the picked data point with mydataset to choose the closest one. Thanks million in advance. best tong __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
so, _if_ 'cd' would be recognized in future releases, it would'nt do any harm, would it? ls() [1] some things here cd(foo) ls() [1] some things here Cue shout of: Hey, it didnt work! ls() is not to /bin/ls what setwd() is to 'cd'. Gets -1 from me. setwd() makes it clear[er] its a working directory on the file system and nothing to do with R's objects. Barry __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. pj On 5/10/06, Prof Brian Ripley [EMAIL PROTECTED] wrote: Exactly. I don't think I have ever used setwd() on Linux. Also, I have never seen this asked before for a Unix-alike, so it seems not to be a FAQ. There is a common tendency for users who run into a problem to think everyone does too, and it isn't necessarily so. Frequently asked questions do make it to the FAQs: it is a defence mechanism for the volunteers supporting R. help.search(directory) gets you there. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Paul E. Johnson Professor, Political Science 1541 Lilac Lane, Room 504 University of Kansas __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] problem rbind after lapply (difference between 2.2.1 and 2.3.0)
Any help on the following would be appreciated... R.version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.1 year 2005 month12 day 20 svn rev 36812 language R set.seed(2) x=matrix(ncol=4, nrow=10) for(i in 1:4) x[,i]=rnorm(10) x=as.data.frame(x) x$V1=c(rep(0,5), rep(1,5)) x V1 V2 V3 V4 1 0 0.41765075 2.090819205 0.7389386 2 0 0.98175278 -1.199925820 0.3189604 3 0 -0.39269536 1.589638200 1.0761644 4 0 -1.03966898 1.954651642 -0.2841577 5 0 1.78222896 0.00493 -0.7766753 6 1 -2.31106908 -2.451706388 -0.5956605 7 1 0.87860458 0.477237303 -1.7259798 8 1 0.03580672 -0.596558169 -0.9025845 9 1 1.01282869 0.792203270 -0.5590619 10 1 0.43226515 0.289636710 -0.2465126 do.call(rbind, lapply(split(x, x$V1), function(x) data.frame(min=min(x$V2), max=max(x$V2 min max 0 -1.039669 1.782229 1 -2.311069 1.012829 yet with the same call on version 2.3.0, I get the following error. do.call(rbind, lapply(split(x, x$V1), function(x) data.frame(min=min(x$V2), max=max(x$V2 Error in data.frame(min = c(-1.039669, -2.311069), max = c(1.782229, : row names contain missing values Any clues on what has changed between the two versions? and a fix to get around this? The actual function (min, max) within the lapply is just an example for illustration only, the issue is the rbinding of results from an lapply call. I've had a look at the changes document packaged with 2.3.0 and found nothing...so wondering whether it's a bug or not? Thanks in advance. Gary __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] problem rbind after lapply (difference between 2.2.1 and 2.3.0)
Update your R: https://www.stat.math.ethz.ch/pipermail/r-devel/2006-May/037542.html On 5/10/06, Gary Collins [EMAIL PROTECTED] wrote: Any help on the following would be appreciated... R.version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.1 year 2005 month12 day 20 svn rev 36812 language R set.seed(2) x=matrix(ncol=4, nrow=10) for(i in 1:4) x[,i]=rnorm(10) x=as.data.frame(x) x$V1=c(rep(0,5), rep(1,5)) x V1 V2 V3 V4 1 0 0.41765075 2.090819205 0.7389386 2 0 0.98175278 -1.199925820 0.3189604 3 0 -0.39269536 1.589638200 1.0761644 4 0 -1.03966898 1.954651642 -0.2841577 5 0 1.78222896 0.00493 -0.7766753 6 1 -2.31106908 -2.451706388 -0.5956605 7 1 0.87860458 0.477237303 -1.7259798 8 1 0.03580672 -0.596558169 -0.9025845 9 1 1.01282869 0.792203270 -0.5590619 10 1 0.43226515 0.289636710 -0.2465126 do.call(rbind, lapply(split(x, x$V1), function(x) data.frame(min=min(x$V2), max=max(x$V2 min max 0 -1.039669 1.782229 1 -2.311069 1.012829 yet with the same call on version 2.3.0, I get the following error. do.call(rbind, lapply(split(x, x$V1), function(x) data.frame(min=min(x$V2), max=max(x$V2 Error in data.frame(min = c(-1.039669, -2.311069), max = c(1.782229, : row names contain missing values Any clues on what has changed between the two versions? and a fix to get around this? The actual function (min, max) within the lapply is just an example for illustration only, the issue is the rbinding of results from an lapply call. I've had a look at the changes document packaged with 2.3.0 and found nothing...so wondering whether it's a bug or not? Thanks in advance. Gary __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
From: Paul Johnson It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. ESS by default will prompt for start-up directory when starting R. Perhaps that was turned off in the lab? Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. Here's one possibility (on WinXP Pro): h - taskCallbackManager() h$add(function(expr, value, ok, visible) { + options(prompt=paste(getwd(), , sep=)) + return(TRUE)}, name=pwdprompt) [1] pwdprompt C:/home setwd(projects) C:/home/projects Andy pj On 5/10/06, Prof Brian Ripley [EMAIL PROTECTED] wrote: Exactly. I don't think I have ever used setwd() on Linux. Also, I have never seen this asked before for a Unix-alike, so it seems not to be a FAQ. There is a common tendency for users who run into a problem to think everyone does too, and it isn't necessarily so. Frequently asked questions do make it to the FAQs: it is a defence mechanism for the volunteers supporting R. help.search(directory) gets you there. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Paul E. Johnson Professor, Political Science 1541 Lilac Lane, Room 504 University of Kansas __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] ape comparative analysis query
I've been comparing variables among objects (taxa) related by known trees, using phylogentically independent contrasts in the ape package, and want to move on to more complex models e.g. by using gls with appropriate correlation terms. My trees contain lots of (hard) polytomies and information about ancestors, which I've been including- creating fully dichotomous trees by using zero branch lengths. For instance: library(ape) tree-read.tree(text=((B1:3,(((D1:5,C1:0):5,B2:0):4,(B3:3,B4:4):0):0):0,A1:0):0;) plot(tree) Where all the B taxa arise in a true polytomy from A1 and B2 is the ancestor of C1 which is the ancestor of D1 I have information on all the taxa and can create independent contrasts for this: x-c(B1=47,D1=43,C1=45,B2=50,B3=47,B4=48,A1=48) y-c(B1=2.9,D1=5.4,C1=2.8,B2=3.5,B3=3.2,B4=3.5,A1=1.8) picx-pic(x,phy=tree) picy-pic(y,phy=tree) Which seems to make sense. However, if I try to use anything more sophisticated I run into problems: dat-as.data.frame(cbind(x,y)) model1-gls(y~x,data=dat, correlation=corBrownian(phy=tree)) Error in corFactor.corStruct(object) : NA/NaN/Inf in foreign function call (arg 1) or equally: model2-compar.gee(y~x,data=dat, phy=tree) [1] Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27 [1] running glm to get initial regression estimate [1] 15.6000 -0.2625 Error in gee(y ~ x, c(1, 1, 1, 1, 1, 1, 1), data = list(x = c(47, 43, : NA/NaN/Inf in foreign function call (arg 14) (NB using the 'data=' argument seems to be necessary in the gls case- if I don't, there is a further problem: 'Row names in dataframe do not match tree tip names. data taken to be in the same order as in tree. in: Initialize.corPhyl(X[[1]], ...)') This seems to go away if I remove the root taxon (A1): tree2-drop.tip(tree, A1) x2-c(B1=47,D1=43,C1=45,B2=50,B3=47,B4=48) y2-c(B1=2.9,D1=5.4,C1=2.8,B2=3.5,B3=3.2,B4=3.5) dat2-as.data.frame(cbind(x2,y2)) model2-gls(y2~x2,data=dat2, correlation=corBrownian(phy=tree2)) This raises various questions: 1) Was I misleading myself that my independent contrasts were valid in the first place? 2) What is it, if anything, about the root taxon that causes this issue, given that other taxa also have zero branch lengths? 3) Is there any way of getting around this and including data on the root taxon, or am I better off just dropping it (ultimately I want to work with much larger trees (up to tens of thousands of taxa) where that one piece of information will become relatively less important) Any help very much appreciated, Chris I'm working with ape 1.8-2 in R 2.1.1 under ubuntu 'Breezy' linux (unfortunately 2.1.1 is the latest easily available in breezy) -- -- Dr Christopher Knight School of Chemistry [EMAIL PROTECTED] The University of Manchester Tel: extn 64414 Faraday Building +44 (0)161 3064414 PO Box 88 Fax: +44 (0)161 3064556 Sackville Street Manchester M60 1QD ` · . ,,(((°UK __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/2006 9:29 AM, Paul Johnson wrote: It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. I think file system directories aren't as central in R as they are in a shell, so it would just be distracting. Most of the time I work in the R workspace, not in the file system. To me the solution is to allow interactive file selection by default, i.e. the default on read.table and similar functions should be file.choose(), rather than having no default and throwing an error. This won't help you in the short run (because file.choose() on Linux isn't all that friendly to beginners), but perhaps it would encourage someone to make it better. file.choose() is quite nice in Windows (and I think on the Mac), so beginners there could be told mydf - read.table() and they'd get something useful. Martin Maechler has disagreed with me about this in the past, but hasn't convinced me that he's right, he's just convinced me that doing nothing is easier than arguing about it. Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/2006 10:11 AM, Liaw, Andy wrote: From: Paul Johnson It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. ESS by default will prompt for start-up directory when starting R. Perhaps that was turned off in the lab? Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. Here's one possibility (on WinXP Pro): h - taskCallbackManager() h$add(function(expr, value, ok, visible) { + options(prompt=paste(getwd(), , sep=)) + return(TRUE)}, name=pwdprompt) [1] pwdprompt C:/home setwd(projects) C:/home/projects That doesn't quite work in the current Rgui, because the menu function to change directories (File | Change dir...) doesn't trigger the callback. Not sure if that's a bug or not... Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 9:29 AM, Paul Johnson wrote: It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. I think file system directories aren't as central in R as they are in a shell, so it would just be distracting. Most of the time I work in the R workspace, not in the file system. To me the solution is to allow interactive file selection by default, i.e. the default on read.table and similar functions should be file.choose(), rather than having no default and throwing an error. This won't help you in the short run (because file.choose() on Linux isn't all that friendly to beginners), but perhaps it would encourage someone to make it better. file.choose() is quite nice in Windows (and I think on the Mac), so beginners there could be told mydf - read.table() and they'd get something useful. Martin Maechler has disagreed with me about this in the past, but hasn't convinced me that he's right, he's just convinced me that doing nothing is easier than arguing about it. I agree with Martin regarding read.table; however, the underlying idea is good and could be achieved via simple wrappers which are the same as the corresponding underlying functions except for the default argument to file: read.table.choose - function(file = file.choose(), ...) read.table(file, ...) read.csv.choose - function(file = file.choose(), ...) read.csv(file, ...) read.delim.choose - function(file = file.choose(), ...) read.delim(file, ...) # test mydata - read.table.choose() in a package available to the users or possibly even in R core. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] new package error message
Thank you very much, Prof. Ripley. This is our own package and in R only, no C or Fortran code. We are doing this on Windows. So it seems that we have to setup the complete R development system for a simple R package to work. Maybe we should just source() the file in, less elegant though. Jason --- Prof Brian Ripley [EMAIL PROTECTED] wrote: In every case I have ever seen mentioned, the solution was to install the package properly by R CMD INSTALL and not by a manual kludge. The message is given if the installation lacks the metadata in the Meta subdirectory which INSTALL creates. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 Jason Liao, http://www.geocities.com/jg_liao Dept. of Biostatistics, http://www2.umdnj.edu/bmtrxweb University of Medicine and Dentistry of New Jersey 683 Hoes Lane West, Piscataway NJ 08854 phone 732-235-5429, School of Public Health office phone 732-235-9824, Cancer Institute of New Jersey office __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/2006 10:45 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 9:29 AM, Paul Johnson wrote: It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. I think file system directories aren't as central in R as they are in a shell, so it would just be distracting. Most of the time I work in the R workspace, not in the file system. To me the solution is to allow interactive file selection by default, i.e. the default on read.table and similar functions should be file.choose(), rather than having no default and throwing an error. This won't help you in the short run (because file.choose() on Linux isn't all that friendly to beginners), but perhaps it would encourage someone to make it better. file.choose() is quite nice in Windows (and I think on the Mac), so beginners there could be told mydf - read.table() and they'd get something useful. Martin Maechler has disagreed with me about this in the past, but hasn't convinced me that he's right, he's just convinced me that doing nothing is easier than arguing about it. I agree with Martin regarding read.table; however, the underlying idea is good and could be achieved via simple wrappers which are the same as the corresponding underlying functions except for the default argument to file: read.table.choose - function(file = file.choose(), ...) read.table(file, ...) read.csv.choose - function(file = file.choose(), ...) read.csv(file, ...) read.delim.choose - function(file = file.choose(), ...) read.delim(file, ...) # test mydata - read.table.choose() in a package available to the users or possibly even in R core. No, I don't think this is a good idea. It would be just as easy to tell people to type read.table(file=file.choose()) with no new package or function necessary. I want the existing basic function to work when used by a beginner in a simple way. What is it that you find objectionable about having a default for the file argument in read.table? I think Martin has said that he doesn't want non-UI functions to be involved with UI functions, but I don't see that: if your code works now, it will be completely unaffected by setting a default for the argument. (Sorry if I summarized the argument incorrectly, Martin, I didn't look it up.) Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] new package error message
On Wed, 10 May 2006, Jason Liao wrote: Thank you very much, Prof. Ripley. This is our own package and in R only, no C or Fortran code. We are doing this on Windows. So it seems that we have to setup the complete R development system for a simple R package to work. Yes, you do (well, you don't need the compilers but you need the tools and perl). We cannot convert help to a viewable format without perl (nor are we likely to in the foreseeable future, help formats having been discussed since 2001). We probably could convert R CMD INSTALL to need only Perl for a package with no C or Fortran code, which might be helpful. Maybe we should just source() the file in, less elegant though. Jason --- Prof Brian Ripley [EMAIL PROTECTED] wrote: In every case I have ever seen mentioned, the solution was to install the package properly by R CMD INSTALL and not by a manual kludge. The message is given if the installation lacks the metadata in the Meta subdirectory which INSTALL creates. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 Jason Liao, http://www.geocities.com/jg_liao Dept. of Biostatistics, http://www2.umdnj.edu/bmtrxweb University of Medicine and Dentistry of New Jersey 683 Hoes Lane West, Piscataway NJ 08854 phone 732-235-5429, School of Public Health office phone 732-235-9824, Cancer Institute of New Jersey office __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 10:45 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 9:29 AM, Paul Johnson wrote: It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. I think file system directories aren't as central in R as they are in a shell, so it would just be distracting. Most of the time I work in the R workspace, not in the file system. To me the solution is to allow interactive file selection by default, i.e. the default on read.table and similar functions should be file.choose(), rather than having no default and throwing an error. This won't help you in the short run (because file.choose() on Linux isn't all that friendly to beginners), but perhaps it would encourage someone to make it better. file.choose() is quite nice in Windows (and I think on the Mac), so beginners there could be told mydf - read.table() and they'd get something useful. Martin Maechler has disagreed with me about this in the past, but hasn't convinced me that he's right, he's just convinced me that doing nothing is easier than arguing about it. I agree with Martin regarding read.table; however, the underlying idea is good and could be achieved via simple wrappers which are the same as the corresponding underlying functions except for the default argument to file: read.table.choose - function(file = file.choose(), ...) read.table(file, ...) read.csv.choose - function(file = file.choose(), ...) read.csv(file, ...) read.delim.choose - function(file = file.choose(), ...) read.delim(file, ...) # test mydata - read.table.choose() in a package available to the users or possibly even in R core. No, I don't think this is a good idea. It would be just as easy to tell people to type read.table(file=file.choose()) with no new package or function necessary. I want the existing basic function to work when used by a beginner in a simple way. I don't think that an idiom of multiple function calls is as simple as issuing a single function call with no args. One possibility is to have a keyword choose on the file function in the same way that file accepts the keyword clipboard. Then one could write: mydata - read.table(choose) I think the wrapper approach is probably preferable though and seems consistent with the way R deals with this in other places such as the ISOdate wrapper. What is it that you find objectionable about having a default for the file argument in read.table? I think Martin has said that he doesn't want non-UI functions to be involved with UI functions, but I don't see that: if your code works now, it will be completely unaffected by setting a default for the argument. (Sorry if I summarized the argument incorrectly, Martin, I didn't look it up.) That would be my objection too. UI should not be tied to the non-UI core. Its basically a loose coupling argument. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] problem rbind after lapply (difference between 2.2.1 and 2.3.0)
This is already fixed in R-patched: o rbind()ing dataframes with a single row could lead to a corrupt data frame (a problem with the fix to PR#8506). On Wed, 10 May 2006, Gary Collins wrote: Any help on the following would be appreciated... R.version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.1 year 2005 month12 day 20 svn rev 36812 language R set.seed(2) x=matrix(ncol=4, nrow=10) for(i in 1:4) x[,i]=rnorm(10) x=as.data.frame(x) x$V1=c(rep(0,5), rep(1,5)) x V1 V2 V3 V4 1 0 0.41765075 2.090819205 0.7389386 2 0 0.98175278 -1.199925820 0.3189604 3 0 -0.39269536 1.589638200 1.0761644 4 0 -1.03966898 1.954651642 -0.2841577 5 0 1.78222896 0.00493 -0.7766753 6 1 -2.31106908 -2.451706388 -0.5956605 7 1 0.87860458 0.477237303 -1.7259798 8 1 0.03580672 -0.596558169 -0.9025845 9 1 1.01282869 0.792203270 -0.5590619 10 1 0.43226515 0.289636710 -0.2465126 do.call(rbind, lapply(split(x, x$V1), function(x) data.frame(min=min(x$V2), max=max(x$V2 min max 0 -1.039669 1.782229 1 -2.311069 1.012829 yet with the same call on version 2.3.0, I get the following error. do.call(rbind, lapply(split(x, x$V1), function(x) data.frame(min=min(x$V2), max=max(x$V2 Error in data.frame(min = c(-1.039669, -2.311069), max = c(1.782229, : row names contain missing values Any clues on what has changed between the two versions? and a fix to get around this? The actual function (min, max) within the lapply is just an example for illustration only, the issue is the rbinding of results from an lapply call. I've had a look at the changes document packaged with 2.3.0 and found nothing...so wondering whether it's a bug or not? Thanks in advance. Gary __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/2006 11:10 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 10:45 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 9:29 AM, Paul Johnson wrote: It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. I think file system directories aren't as central in R as they are in a shell, so it would just be distracting. Most of the time I work in the R workspace, not in the file system. To me the solution is to allow interactive file selection by default, i.e. the default on read.table and similar functions should be file.choose(), rather than having no default and throwing an error. This won't help you in the short run (because file.choose() on Linux isn't all that friendly to beginners), but perhaps it would encourage someone to make it better. file.choose() is quite nice in Windows (and I think on the Mac), so beginners there could be told mydf - read.table() and they'd get something useful. Martin Maechler has disagreed with me about this in the past, but hasn't convinced me that he's right, he's just convinced me that doing nothing is easier than arguing about it. I agree with Martin regarding read.table; however, the underlying idea is good and could be achieved via simple wrappers which are the same as the corresponding underlying functions except for the default argument to file: read.table.choose - function(file = file.choose(), ...) read.table(file, ...) read.csv.choose - function(file = file.choose(), ...) read.csv(file, ...) read.delim.choose - function(file = file.choose(), ...) read.delim(file, ...) # test mydata - read.table.choose() in a package available to the users or possibly even in R core. No, I don't think this is a good idea. It would be just as easy to tell people to type read.table(file=file.choose()) with no new package or function necessary. I want the existing basic function to work when used by a beginner in a simple way. I don't think that an idiom of multiple function calls is as simple as issuing a single function call with no args. One possibility is to have a keyword choose on the file function in the same way that file accepts the keyword clipboard. Then one could write: mydata - read.table(choose) I think the wrapper approach is probably preferable though and seems consistent with the way R deals with this in other places such as the ISOdate wrapper. What is it that you find objectionable about having a default for the file argument in read.table? I think Martin has said that he doesn't want non-UI functions to be involved with UI functions, but I don't see that: if your code works now, it will be completely unaffected by setting a default for the argument. (Sorry if I summarized the argument incorrectly, Martin, I didn't look it up.) That would be my objection too. UI should not be tied to the non-UI core. Its basically a loose coupling argument. I don't accept that argument, because in R everything* is interactive. There isn't a non-UI core. The function arguments are part of the user interface. Duncan Murdoch * Well, maybe this should be restricted to functions visible to the user, but everything we've been discussing so far is visible and is part of the user interface. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] problems with optimize (again)
Can someone please explain what the $minimum result of the optimize function actually is? I'm trying to optimize the function: fitIT-function(ampFac,ts_wave1,ts_template){ template-stretchWaveTime(ts_template,ampFac) fit-calcFit(ts_wave1,template) return(fit) } with optimize(f=fitIT,interval=c(0.5,4),ts_wave1=test.data[,1],ts_template=test.data[,1]) $minimum [1] 3.764685 $objective [1] 1.037864 however when I run calcFit(test.data[,1],stretchWaveTime(test.data[,1],1.037864)) [1] 0.3059954 I thought $minimum and the result of the calcFit function should be the same. As a further question I'm going to need to nest optimize functions so I can optimize over two variables. How can I get the minimum and objective back from the nested function f, as well as the value to be minimised? I hope this makes sense and I apologise for the lack of reproducible code. If it will help I can include that in a further e-mail. Thanks Tom __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] importing a list
Hi, all. I'm trying to automate some regression operations in R leaving the possibility to modify the predictors in the regression. For example, I have saved in a list the results and then exported as a txt file, in which we can modify the predictors, putting for example lm(y~x^2) instead of having lm(y~x) as in the original model. Now, I need to import in R the txt file as a list to evaluate the model. Is that possible? I played around with source() and file() but can't figure it out. Thanks. Grazia M. Grazia Pittau, Ph.D. Post-Doctoral Research Fellow Department of Statistics Columbia University 1255 Amsterdam Avenue New York, NY 10027 [EMAIL PROTECTED] Phone: 212.851.2160 Fax: 212.851.2164 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On Wed, May 10, 2006 at 11:26:55AM -0400, Duncan Murdoch wrote: On 5/10/2006 11:10 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: What is it that you find objectionable about having a default for the file argument in read.table? I think Martin has said that he doesn't want non-UI functions to be involved with UI functions, but I don't see that: if your code works now, it will be completely unaffected by setting a default for the argument. (Sorry if I summarized the argument incorrectly, Martin, I didn't look it up.) That would be my objection too. UI should not be tied to the non-UI core. Its basically a loose coupling argument. I don't accept that argument, because in R everything* is interactive. There isn't a non-UI core. The function arguments are part of the user interface. It seems to me that there might be a misunderstanding here; as the term user is used to refer to a person interacting with the computer on the one hand, and to refer to a programmer using R on the other hand. Everything being part of the user interface, in the sense of every user-visible function being part of the API, does not and should not imply that everything should be interactive. In my experience, interactivity is a rather double-edged thing: On the one hand, it facilitates learning and exploration, but on the other hand, its improper use is frequently detrimental to reproducibility of scientific computation. Best regards, Jan -- +- Jan T. Kim ---+ | email: [EMAIL PROTECTED] | | WWW: http://www.cmp.uea.ac.uk/people/jtk | *-= hierarchical systems are for files, not for humans =-* __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 11:10 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 10:45 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: On 5/10/2006 9:29 AM, Paul Johnson wrote: It is a FAQ in our Linux lab. People start emacs and fire up R via ess, and then they have no idea 'where they are. For computer experts, it is not a problem, but for people who don't know much about computers, it is a pretty big problem. They have data in some subdirectory, but almost invariably they don't get emacs R started from that same place. Unfortunately, for our users, it does not help to simply re-label setwd as cd. Both commands imply a deeper understanding of the OS than they have. Also, unfortunately, these are the same people who don't understand that FAQs exist and should be consulted. These people are so new/timid that asking in r-help would be the last thing to cross their mind. I've wondered if it would not help to have the R prompt include the directory name, as in an x terminal. I think file system directories aren't as central in R as they are in a shell, so it would just be distracting. Most of the time I work in the R workspace, not in the file system. To me the solution is to allow interactive file selection by default, i.e. the default on read.table and similar functions should be file.choose(), rather than having no default and throwing an error. This won't help you in the short run (because file.choose() on Linux isn't all that friendly to beginners), but perhaps it would encourage someone to make it better. file.choose() is quite nice in Windows (and I think on the Mac), so beginners there could be told mydf - read.table() and they'd get something useful. Martin Maechler has disagreed with me about this in the past, but hasn't convinced me that he's right, he's just convinced me that doing nothing is easier than arguing about it. I agree with Martin regarding read.table; however, the underlying idea is good and could be achieved via simple wrappers which are the same as the corresponding underlying functions except for the default argument to file: read.table.choose - function(file = file.choose(), ...) read.table(file, ...) read.csv.choose - function(file = file.choose(), ...) read.csv(file, ...) read.delim.choose - function(file = file.choose(), ...) read.delim(file, ...) # test mydata - read.table.choose() in a package available to the users or possibly even in R core. No, I don't think this is a good idea. It would be just as easy to tell people to type read.table(file=file.choose()) with no new package or function necessary. I want the existing basic function to work when used by a beginner in a simple way. I don't think that an idiom of multiple function calls is as simple as issuing a single function call with no args. One possibility is to have a keyword choose on the file function in the same way that file accepts the keyword clipboard. Then one could write: mydata - read.table(choose) I think the wrapper approach is probably preferable though and seems consistent with the way R deals with this in other places such as the ISOdate wrapper. What is it that you find objectionable about having a default for the file argument in read.table? I think Martin has said that he doesn't want non-UI functions to be involved with UI functions, but I don't see that: if your code works now, it will be completely unaffected by setting a default for the argument. (Sorry if I summarized the argument incorrectly, Martin, I didn't look it up.) That would be my objection too. UI should not be tied to the non-UI core. Its basically a loose coupling argument. I don't accept that argument, because in R everything* is interactive. There isn't a non-UI core. The function arguments are part of the user interface. Duncan Murdoch * Well, maybe this should be restricted to functions visible to the user, but everything we've been discussing so far is visible and is part of the user interface. Another idea is to implement a menu item in the Windows GUI that does a read.table(file.choose(), ...). __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] importing a list
?dump ?source But do you really need to save the fitted object as a txt file? Saving/loading it in native format (?save ?load) and then using ?update would seem to be more straightforward. -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA The business of the statistician is to catalyze the scientific learning process. - George E. P. Box -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Wednesday, May 10, 2006 8:24 AM To: R-help@stat.math.ethz.ch Subject: [R] importing a list Hi, all. I'm trying to automate some regression operations in R leaving the possibility to modify the predictors in the regression. For example, I have saved in a list the results and then exported as a txt file, in which we can modify the predictors, putting for example lm(y~x^2) instead of having lm(y~x) as in the original model. Now, I need to import in R the txt file as a list to evaluate the model. Is that possible? I played around with source() and file() but can't figure it out. Thanks. Grazia M. Grazia Pittau, Ph.D. Post-Doctoral Research Fellow Department of Statistics Columbia University 1255 Amsterdam Avenue New York, NY 10027 [EMAIL PROTECTED] Phone: 212.851.2160 Fax: 212.851.2164 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] problems with optimize (again)
you should use: optmz - optimize(fitIT, interval = c(0.5, 4), ts_wave1 = test.data[, 1], ts_template = test.data[, 1]) calcFit(test.data[, 1], stretchWaveTime(test.data[, 1], optmz$minimum) which should give the same as optmz$objective, i.e., the optmz$minimum is the value of 'ampFac' argument for which fitIT() attains its minimum value; the value of fitIT() at optmz$minimum is optmz$objective. I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: tom wright [EMAIL PROTECTED] To: R-Stat Help R-help@stat.math.ethz.ch Sent: Wednesday, May 10, 2006 2:18 PM Subject: [R] problems with optimize (again) Can someone please explain what the $minimum result of the optimize function actually is? I'm trying to optimize the function: fitIT-function(ampFac,ts_wave1,ts_template){ template-stretchWaveTime(ts_template,ampFac) fit-calcFit(ts_wave1,template) return(fit) } with optimize(f=fitIT,interval=c(0.5,4),ts_wave1=test.data[,1],ts_template=test.data[,1]) $minimum [1] 3.764685 $objective [1] 1.037864 however when I run calcFit(test.data[,1],stretchWaveTime(test.data[,1],1.037864)) [1] 0.3059954 I thought $minimum and the result of the calcFit function should be the same. As a further question I'm going to need to nest optimize functions so I can optimize over two variables. How can I get the minimum and objective back from the nested function f, as well as the value to be minimised? I hope this makes sense and I apologise for the lack of reproducible code. If it will help I can include that in a further e-mail. Thanks Tom __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] problems with optimize (again)
tom wright wrote: Can someone please explain what the $minimum result of the optimize function actually is? I'm trying to optimize the function: fitIT-function(ampFac,ts_wave1,ts_template){ template-stretchWaveTime(ts_template,ampFac) fit-calcFit(ts_wave1,template) return(fit) } with optimize(f=fitIT,interval=c(0.5,4),ts_wave1=test.data[,1],ts_template=test.data[,1]) $minimum [1] 3.764685 $objective [1] 1.037864 however when I run calcFit(test.data[,1],stretchWaveTime(test.data[,1],1.037864)) [1] 0.3059954 What is calcFit? I thought $minimum and the result of the calcFit function should be the same. Why should to different algorithms (at least I guess so, since you don't say what calcFit() does) find the same *local* minimum? As a further question I'm going to need to nest optimize functions so I can optimize over two variables. How can I get the minimum and objective back from the nested function f, as well as the value to be minimised? Use optim(). Uwe Ligges I hope this makes sense and I apologise for the lack of reproducible code. If it will help I can include that in a further e-mail. Thanks Tom __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] importing a list
Hello Grazia, On May 10, 2006, at 10:23 AM, [EMAIL PROTECTED] wrote: I'm trying to automate some regression operations in R leaving the possibility to modify the predictors in the regression. For example, I have saved in a list the results and then exported as a txt file, in which we can modify the predictors, putting for example lm(y~x^2) instead of having lm(y~x) as in the original model. Now, I need to import in R the txt file as a list to evaluate the model. Is that possible? I played around with source() and file() but can't figure it out. What OS are you using? When you say the list is in a text file, how is the text file structured, e.g. CSV, TSV, fixed-width, other? Here's an example of creating, writing, and reading a data.frame: d - data.frame(cbind(x=1, y=1:10)) write.table(d, file=r-data.txt, sep=\t, na=, quote=FALSE, row.names = FALSE) d.new - read.delim(r-data.txt) HTH. Regards, - Robert http://www.cwelug.org/downloads Help others get OpenSource software. Distribute FLOSS for Windows, Linux, *BSD, and MacOS X with BitTorrent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Batch printing of existing postscript files with file names included
Dear R users, I have created a series of postscript files and I´d like to print them with the file name added to the printout. Is there a way of reading these files into R (e.g. using rimage after conversion to jpeg), adding the file name, and then sending the files to a windows printer? I have already tried the rimage package, and several image batch conversion program, but none was so far able to do the job, unfortunately. I´m using R 2.3.0 on Windows XP. Thank you very much for your help! Best regards Christoph __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On 5/10/2006 12:15 PM, Jan T. Kim wrote: On Wed, May 10, 2006 at 11:26:55AM -0400, Duncan Murdoch wrote: On 5/10/2006 11:10 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: What is it that you find objectionable about having a default for the file argument in read.table? I think Martin has said that he doesn't want non-UI functions to be involved with UI functions, but I don't see that: if your code works now, it will be completely unaffected by setting a default for the argument. (Sorry if I summarized the argument incorrectly, Martin, I didn't look it up.) That would be my objection too. UI should not be tied to the non-UI core. Its basically a loose coupling argument. I don't accept that argument, because in R everything* is interactive. There isn't a non-UI core. The function arguments are part of the user interface. It seems to me that there might be a misunderstanding here; as the term user is used to refer to a person interacting with the computer on the one hand, and to refer to a programmer using R on the other hand. One of the design goals of S and R is to blur the distinction between users and programmers. It is a continuum. R is designed to gently urge non-programmers to become programmers, because the designers think that's the way statistical computing should be done. Everything being part of the user interface, in the sense of every user-visible function being part of the API, does not and should not imply that everything should be interactive. No, I didn't suggest that. What I was suggesting is that it should be *convenient* to use read.table interactively, not that it should be required. (It's already possible, but not convenient, especially for a beginner who doesn't know the secret incantation.) In my experience, interactivity is a rather double-edged thing: On the one hand, it facilitates learning and exploration, but on the other hand, its improper use is frequently detrimental to reproducibility of scientific computation. I definitely agree with that. It should be convenient to use R non-interactively as well. Anyone who wants reproducibility should be writing packages and scripts or vignettes that run non-interactively. That's why I am emphasizing that this change will have no effect on existing code. I wouldn't suggest it if it did. Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] finding centroids of clusters created with hclust
Replying to myself for the record: Moritz Lennert wrote: Hello, Can someone point me to documentation or ideas on how to calculate the centroids of clusters identified with hclust ? I would like to be able to chose the number of clusters (in the style of cutree) and then get the centroids of these clusters. This seems like a quite obvious task to me, but I haven't been able to put my hands on a relevant command. Here's a simple function that does the job for me: Variables: data: matrix of original (absolute value) data introduced into hclust or HierClust clust: result of a 'cutree' call on the results of the hclust or HierClust call Value: a matrix of relative values of the variables at the centroids of the types function (data, clust) { nvars=length(data[1,]) ntypes=max(clust) centroids-matrix(0,ncol=nvars,nrow=ntypes) for(i in 1:ntypes) { c-rep(0,nvars) n-0 for(j in names(clust[clust==i])) { n-n+1 c-c+data[j,] } centroids[i,]-c/n } rownames(centroids)-c(1:ntypes) colnames(centroids)-colnames(data) centroids } Moritz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Until the key pressed: FOR-LOOP-Clustering
Not sure exactly what you're asking, but I'll take a stab at it: Try adding par(ask=TRUE) before the for() loop. Andy From: j.joshua thomas Hi All, These are the code that i used to plot the kmeans clustering. DataSetS01022-rbind(matrix(rnorm(50),ncol=2), + matrix(rnorm(50),ncol=2)) colnames(DataSetS01022) -c(timeslot,em) (cl -kmeans(DataSetS01022,2)) plot (DataSetS01022, col=cl$cluster) points(cl$centers, col = 1:2, pch = 8, cex=2) Now, i would like to do *the same process for 5 times*, so that i can view the clusters in data have been moving and the center changes with various results. (Animation-like) My main concern here, is how to put this in a *loop form* or any other methods are available? I tried with:* clarax$clusinfo I thought of doing it to activate Device (1) (2)* ** *Really need anyones help!* ** *JJ* *---* -- Lecturer J. Joshua Thomas KDU College Penang Campus Research Student, University Sains Malaysia [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] importing a list
Robert, thanks first of all. Actually, the txt file I want to read as a list is like that: $y $y$mod.1 [1] impute.1.y=glm(I(y0)~x+ z+ w,family=binomial(link =logit),data= data) $y$mod.2 [1] impute.2.y=lm(I(log (y))~x+ z+ w,subset=y0) $x [1] impute.x=lm(x~y+ z+ w, data= data) $z [1] impute.z=lm(z~y+ x+ w,data= data) that is a list of 4 elements with the first element a list as well. To save the file I'm using this function: list2ascii - function(x, file=paste(deparse(substitute(x)),.txt,sep=)) { tmp.wid = getOption(width) options(width=1) sink(file) print(x) sink() options(width=tmp.wid) return(invisible(NULL)) } So, I should read it using read.fwf(models.txt, width=getOption(width)) The problem is that I'm not getting back a list! Probably I have to play with write.table?!! thanks, grazia On Wed, 10 May 2006, Robert Citek wrote: Hello Grazia, On May 10, 2006, at 10:23 AM, [EMAIL PROTECTED] wrote: I'm trying to automate some regression operations in R leaving the possibility to modify the predictors in the regression. For example, I have saved in a list the results and then exported as a txt file, in which we can modify the predictors, putting for example lm(y~x^2) instead of having lm(y~x) as in the original model. Now, I need to import in R the txt file as a list to evaluate the model. Is that possible? I played around with source() and file() but can't figure it out. What OS are you using? When you say the list is in a text file, how is the text file structured, e.g. CSV, TSV, fixed-width, other? Here's an example of creating, writing, and reading a data.frame: d - data.frame(cbind(x=1, y=1:10)) write.table(d, file=r-data.txt, sep=\t, na=, quote=FALSE, row.names = FALSE) d.new - read.delim(r-data.txt) HTH. Regards, - Robert http://www.cwelug.org/downloads Help others get OpenSource software. Distribute FLOSS for Windows, Linux, *BSD, and MacOS X with BitTorrent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] attach() warning message
Hi, I am using R 2.3.0 and my .First() file has the following line which causes a warning: attach(what=c:/r/jmb/.RData, pos=3) I get the following message after .First() is loaded: During startup - Warning message: use of NULL environment is deprecated Is there a simple fix that will eliminate this warning message? Thanks. J. Breiwick __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] identify high dimension data point
The tkBrush function in the TeachingDemos package may help. If you just want to see where a selected point apears in each of the pairwise plots based on selecting it in 1 plot then this function will do that for you (note it depends on having installed the tcltk package as well). If you need the index of a selected point then you can do something like: out1 - tkBrush(mydata) # now select the point of interest, click to make the change persistent. which(out1$col != 'black') # this gives the observation (row) number Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tong Wang Sent: Wednesday, May 10, 2006 5:05 AM To: R help Subject: [R] identify high dimension data point Hi there, I am having some trouble with the Identify() function, it looks that the identify() function only works on the plot of 2-dimension data set. In high-dimension case , if I use pairs() + identify() in hope to observe all the components of the data point I selected, I get error message. Is there anyway to get around this ? Plus, I also need it to actually compare all the components of the picked data point with mydataset to choose the closest one. Thanks million in advance. best tong __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] finding centroids of clusters created with hclust
On Wed, 2006-05-10 at 18:59 +0200, Moritz Lennert wrote: Replying to myself for the record: Moritz Lennert wrote: Hello, Can someone point me to documentation or ideas on how to calculate the centroids of clusters identified with hclust ? I would like to be able to chose the number of clusters (in the style of cutree) and then get the centroids of these clusters. This seems like a quite obvious task to me, but I haven't been able to put my hands on a relevant command. Sorry, Moritz, I meant to reply to your original post, but deleted it from my emailer accidentally and hadn't had chance to use the archives to follow up. Anyway, Venables and Ripley's Modern Applied Statistics with S (4th Ed) [and earlier editions - it is in my 3rd Edition for example] has an example of doing what you want to do on page 318 of the 4th Edition. They use the centre's of the hclust results as starting points for a k-means, so we only need the preliminary bits of their example: library(MASS) swiss.x - as.matrix(swiss) h - hclust(dist(swiss.x), method = average) initial - tapply(swiss.x, list(rep(cutree(h, 3), ncol(swiss.x)), col(swiss.x)), mean) dimnames(initial) - list(NULL, dimnames(swiss.x)[[2]]) initial Which gives almost the same output as your function: fun - function (data, clust) { nvars=length(data[1,]) ntypes=max(clust) centroids-matrix(0,ncol=nvars,nrow=ntypes) for(i in 1:ntypes) { c-rep(0,nvars) n-0 for(j in names(clust[clust==i])) { n-n+1 c-c+data[j,] } centroids[i,]-c/n } rownames(centroids)-c(1:ntypes) colnames(centroids)-colnames(data) centroids } fun(swiss.x, cutree(h, 3)) Wrapping the Venables Ripley version into a function to give the same output as your function: ## ## clust.means - function to find centroids of clusters ## based on example by Venables Ripley, MASS 4thEd, Page 318 [1] ## ## x= input data as data.frame or matrix ## res.clust= object of class hclust ## groups = number of groups to cut dendrogram into ## ## References: ## ## [1] Venables, W.N. and Ripley, B.D. (2002) Modern Applied Statistics ## with S. 4th Edition. Springer. clust.means - function(x, res.clust, groups) { if(!is.matrix(x)) x - as.matrix(x) means - tapply(x, list(rep(cutree(res.clust, groups), ncol(x)), col(x)), mean) dimnames(means) - list(NULL, dimnames(x)[[2]]) return(as.data.frame(means)) } clust.means(swiss, h, 3) Your function is faster here: system.time(for(i in 1:1) fun(swiss.x, cutree(h, 3))) [1] 8.917 0.000 9.695 0.000 0.000 system.time(for(i in 1:1) clust.means(swiss, h, 3)) [1] 31.642 0.008 35.348 0.000 0.000 But I think the example is instructive about using R. Sometimes vectorisation can make a big time saving over a loop - here it doesn't. HTH G Here's a simple function that does the job for me: Variables: data: matrix of original (absolute value) data introduced into hclust or HierClust clust: result of a 'cutree' call on the results of the hclust or HierClust call Value: a matrix of relative values of the variables at the centroids of the types function (data, clust) { nvars=length(data[1,]) ntypes=max(clust) centroids-matrix(0,ncol=nvars,nrow=ntypes) for(i in 1:ntypes) { c-rep(0,nvars) n-0 for(j in names(clust[clust==i])) { n-n+1 c-c+data[j,] } centroids[i,]-c/n } rownames(centroids)-c(1:ntypes) colnames(centroids)-colnames(data) centroids } Moritz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% * Note new Address, Telephone Fax numbers from 6th April 2006 * %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson ECRC ENSIS [t] +44 (0)20 7679 0522 UCL Department of Geography [f] +44 (0)20 7679 0565 Pearson Building [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. [w] http://www.ucl.ac.uk/~ucfagls/ WC1E 6BT. %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] importing a list
A search for the list2ascii() function, led me to this post by Mike Prager: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/66335.html in which Mike specifically notes: To write it to a file that can be read by R, I would suggest using dput instead. Thus, instead of using list2ascii() on your list object, use: sink(YourTextFile) dput(YourListObject) sink() You can then re-read the text file back into R using: source(YourTextFile) provided of course that you don't compromise the integrity of the file content when you edit the model formulae. See ?sink (or ?capture.output) and ?dput for more help here. However, I wholeheartedly support Bert's recommendation that you use save()/load()/update(), since these functions enable you to work with R objects directly in their native format _within R_, as opposed to messing around with textual representations in an external editor where you risk introducing errors through typos. R is an incredibly powerful environment, when you work within it to take advantage of its power, rather then trying to work around it to obviate it. HTH, Marc Schwartz On Wed, 2006-05-10 at 13:05 -0400, [EMAIL PROTECTED] wrote: Robert, thanks first of all. Actually, the txt file I want to read as a list is like that: $y $y$mod.1 [1] impute.1.y=glm(I(y0)~x+ z+ w,family=binomial(link =logit),data= data) $y$mod.2 [1] impute.2.y=lm(I(log (y))~x+ z+ w,subset=y0) $x [1] impute.x=lm(x~y+ z+ w, data= data) $z [1] impute.z=lm(z~y+ x+ w,data= data) that is a list of 4 elements with the first element a list as well. To save the file I'm using this function: list2ascii - function(x, file=paste(deparse(substitute(x)),.txt,sep=)) { tmp.wid = getOption(width) options(width=1) sink(file) print(x) sink() options(width=tmp.wid) return(invisible(NULL)) } So, I should read it using read.fwf(models.txt, width=getOption(width)) The problem is that I'm not getting back a list! Probably I have to play with write.table?!! thanks, grazia On Wed, 10 May 2006, Robert Citek wrote: Hello Grazia, On May 10, 2006, at 10:23 AM, [EMAIL PROTECTED] wrote: I'm trying to automate some regression operations in R leaving the possibility to modify the predictors in the regression. For example, I have saved in a list the results and then exported as a txt file, in which we can modify the predictors, putting for example lm(y~x^2) instead of having lm(y~x) as in the original model. Now, I need to import in R the txt file as a list to evaluate the model. Is that possible? I played around with source() and file() but can't figure it out. What OS are you using? When you say the list is in a text file, how is the text file structured, e.g. CSV, TSV, fixed-width, other? Here's an example of creating, writing, and reading a data.frame: d - data.frame(cbind(x=1, y=1:10)) write.table(d, file=r-data.txt, sep=\t, na=, quote=FALSE, row.names = FALSE) d.new - read.delim(r-data.txt) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] importing a list
On Wed, 2006-05-10 at 13:19 -0500, Marc Schwartz (via MN) wrote: A search for the list2ascii() function, led me to this post by Mike Prager: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/66335.html in which Mike specifically notes: To write it to a file that can be read by R, I would suggest using dput instead. Thus, instead of using list2ascii() on your list object, use: sink(YourTextFile) dput(YourListObject) sink() You can then re-read the text file back into R using: source(YourTextFile) Actually, quick correction here. That should be: NewListObject - dget(YourTextFile) dget() will enable simple assignment of the textual input, whereas source() will just evaluate it and assignment with source() creates a list object, unless you use: NewListObject - source(YourTextFile)$value Marc __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] attach() warning message
On 5/10/2006 2:02 PM, J.M. Breiwick wrote: Hi, I am using R 2.3.0 and my .First() file has the following line which causes a warning: attach(what=c:/r/jmb/.RData, pos=3) I get the following message after .First() is loaded: During startup - Warning message: use of NULL environment is deprecated Is there a simple fix that will eliminate this warning message? Thanks. This probably means you have some binary objects in your .RData which were produced under an earlier version of R and will not be compatible with R 2.4.x. I think it will continue to make automatic conversions (we strive to be able to read old binary versions), but the automatic conversions might not be correct. There are two things to do: The simplest one is to ignore the warning, save your data from R 2.3.0, and hope that the automatic corrections were successful. I recommend you backup the old file first, and test carefully afterwards. A better solution is to load your .RData in the version of R that was used to create it, dump it in text format (using the dump() function), and then reload it into R 2.3.0 using source(). It is possible that this will create very small changes to some numerical values, since I don't think we guarantee that the text representation is always parsed to identical values. If you've got a mixture of old and new items (because you have added new things), you may not be able to do this: the old version of R may not be able to read the file. In that case you can try dump() from R 2.3.0, hoping that the automatic fixups were correct; you can look over the file (if it's not too large) to check. In future, I'd recommend that you don't keep things in binary format; R changes, and you might run into this sort of problem again. We try to keep backwards compatibility, but it's not always possible. Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] importing a list
Marc, it's perfect THANKS a lot for your help. grazia On Wed, 10 May 2006, Marc Schwartz (via MN) wrote: On Wed, 2006-05-10 at 13:19 -0500, Marc Schwartz (via MN) wrote: A search for the list2ascii() function, led me to this post by Mike Prager: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/66335.html in which Mike specifically notes: To write it to a file that can be read by R, I would suggest using dput instead. Thus, instead of using list2ascii() on your list object, use: sink(YourTextFile) dput(YourListObject) sink() You can then re-read the text file back into R using: source(YourTextFile) Actually, quick correction here. That should be: NewListObject - dget(YourTextFile) dget() will enable simple assignment of the textual input, whereas source() will just evaluate it and assignment with source() creates a list object, unless you use: NewListObject - source(YourTextFile)$value Marc __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] command completion?
Does R have command or object name completion? R has several functions of modern shells built into it e.g. history mechanism, command-line editing, emacs key bindings. However, it doesn't seem to have command or object name completion. For example, in bash I can begin typing a command and then press the tab key to have the shell fill in the rest of the command (works for environment variables and filenames, too). Does command completion exist and I'm not using it correctly or don't have it turned on, or is it really not there? Regards, - Robert http://www.cwelug.org/downloads Help others get OpenSource software. Distribute FLOSS for Windows, Linux, *BSD, and MacOS X with BitTorrent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] integer vector to a string
hi there is there a way that i can coerce a vector of integers to ONE string with the numbers comma separated like: 1:500 -1,2,3, ..., 500 i've tried deparse, but it divides the result into a vector of string (depending on the setting of width with a max of 500) regards soren __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] command completion?
On 5/10/2006 2:52 PM, Robert Citek wrote: Does R have command or object name completion? R has several functions of modern shells built into it e.g. history mechanism, command-line editing, emacs key bindings. However, it doesn't seem to have command or object name completion. For example, in bash I can begin typing a command and then press the tab key to have the shell fill in the rest of the command (works for environment variables and filenames, too). Does command completion exist and I'm not using it correctly or don't have it turned on, or is it really not there? It's not there, because some of the shells that run R supply it. I don't use one, but I think a couple of the popular ones are ESS and the Mac OS GUI. It would be a nice addition to the Windows GUI, but I don't know anyone who is working on it. I don't think it would be a welcome change to the console versions; some of them use readline's filename completion which would almost certainly be broken by this. Others need to run under ESS. Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] integer vector to a string
On 5/10/2006 2:49 PM, Søren Merser wrote: hi there is there a way that i can coerce a vector of integers to ONE string with the numbers comma separated like: 1:500 -1,2,3, ..., 500 i've tried deparse, but it divides the result into a vector of string (depending on the setting of width with a max of 500) paste(1:500, collapse=,) Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] integer vector to a string
Søren Merser [EMAIL PROTECTED] writes: hi there is there a way that i can coerce a vector of integers to ONE string with the numbers comma separated like: 1:500 -1,2,3, ..., 500 i've tried deparse, but it divides the result into a vector of string (depending on the setting of width with a max of 500) paste(1:500, collapse=,) -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] integer vector to a string
probably you need ?paste, e.g., paste(1:500, collapse = ,) I hope it helps. Best, Dimitris Quoting Søren Merser [EMAIL PROTECTED]: hi there is there a way that i can coerce a vector of integers to ONE string with the numbers comma separated like: 1:500 -1,2,3, ..., 500 i've tried deparse, but it divides the result into a vector of string (depending on the setting of width with a max of 500) regards soren __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On Wed, May 10, 2006 at 12:56:23PM -0400, Duncan Murdoch wrote: On 5/10/2006 12:15 PM, Jan T. Kim wrote: On Wed, May 10, 2006 at 11:26:55AM -0400, Duncan Murdoch wrote: On 5/10/2006 11:10 AM, Gabor Grothendieck wrote: On 5/10/06, Duncan Murdoch [EMAIL PROTECTED] wrote: What is it that you find objectionable about having a default for the file argument in read.table? I think Martin has said that he doesn't want non-UI functions to be involved with UI functions, but I don't see that: if your code works now, it will be completely unaffected by setting a default for the argument. (Sorry if I summarized the argument incorrectly, Martin, I didn't look it up.) That would be my objection too. UI should not be tied to the non-UI core. Its basically a loose coupling argument. I don't accept that argument, because in R everything* is interactive. There isn't a non-UI core. The function arguments are part of the user interface. It seems to me that there might be a misunderstanding here; as the term user is used to refer to a person interacting with the computer on the one hand, and to refer to a programmer using R on the other hand. One of the design goals of S and R is to blur the distinction between users and programmers. It is a continuum. R is designed to gently urge non-programmers to become programmers, because the designers think that's the way statistical computing should be done. That's an idea I like very much too -- much better than the currently popular idea of protecting users from the unfriendliness of programming, anyway... Everything being part of the user interface, in the sense of every user-visible function being part of the API, does not and should not imply that everything should be interactive. No, I didn't suggest that. What I was suggesting is that it should be *convenient* to use read.table interactively, not that it should be required. (It's already possible, but not convenient, especially for a beginner who doesn't know the secret incantation.) Well, not knowing a secret is always inconvenient... ;-) In my experience, interactivity is a rather double-edged thing: On the one hand, it facilitates learning and exploration, but on the other hand, its improper use is frequently detrimental to reproducibility of scientific computation. I definitely agree with that. It should be convenient to use R non-interactively as well. Anyone who wants reproducibility should be writing packages and scripts or vignettes that run non-interactively. Ok, I fully agree with this -- seems that I've interpreted the statement that in R everything is interactive a bit too narrowly. That's why I am emphasizing that this change will have no effect on existing code. I wouldn't suggest it if it did. That's an important point too, obviously. I'm not entirely convinced about the convenience aspect, as I find file choosers of all sorts disruptive to workflow... but that's perhaps a matter of personal taste. Best regards, Jan -- +- Jan T. Kim ---+ | email: [EMAIL PROTECTED] | | WWW: http://www.cmp.uea.ac.uk/people/jtk | *-= hierarchical systems are for files, not for humans =-* __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Courses - R/Splus Advanced Programming ***by the R Development Core Tean Guru In Seattle and San Francisco ***June/July***
XLSolutions Corporation ([1]www.xlsolutions-corp.com) is proud to announce our*** R/S Advanced Programming ***courses in Seattle and San Francisco this summer by the R Development Core Team Guru! (1) R/Splus Advanced Programming *** Seattle/June 8-9, 2006 *** *** San Francisco / July 13-14, 2006 *** [2]http://www.xlsolutions-corp.com/Radv.htm (2) R/Splus Fundamentals and Programming Techniques *** Boston / June 29-30, 2006*** *** Seattle/June 5-6, 2006 *** [3]http://www.xlsolutions-corp.com/Rfund.htm Ask for group discount and reserve your seat Now - Earlybird Rates. Payment due after theclass!EmailSueTurner: [EMAIL PROTECTED] (1) R/Splus Advanced Programming Course Outline: - Overview of R/S fundamentals: Syntax and Semantics - Class and Inheritance in R/S-Plus - Concepts, Construction and good use of language objects - Coercion and efficiency - Object-oriented programming in R and S-Plus - Advanced manipulation tools: Parse, Deparse, Substitute, etc. - How to fully take advantage of Vectorization - Generic and Method Functions - Search path, databases and frames Visibility - Working with large objects - Handling Properly Recursion and iterative calculations - Managing loops; For (S-Plus) and for() loops - Consequences of Lazy Evaluation - Efficient Code practices for large computations - Memory management and Resource monitoring - Writing R/S-Plus functions to call compiled code - Writing and debugging compiled code for R/S-Plus system - Connecting R/S-Plus to External Data Sources - Understanding the structure of model fitting functions in R/S-Plus - Designing and Packaging efficiently a new model function It'll also deal with lots of S-Plus efficiency issues and any special topics from participants is welcome. (2) R/Splus Fundamentals and Programming Techniques Course outline. - An Overview of R and S - Data Manipulation and Graphics - Using Lattice Graphics - A Comparison of R and S-Plus - How can R Complement SAS? - Writing Functions - Avoiding Loops - Vectorization - Statistical Modeling - Project Management - Techniques for Effective use of R and S - Enhancing Plots - Using High-level Plotting Functions - Building and Distributing Packages (libraries) - Connecting; ODBC, Rweb, Orca via sockets and via Rjava Email us for group discounts. Email Sue Turner: [EMAIL PROTECTED] Phone: 206-686-1578 Visit us: [6]www.xlsolutions-corp.com/training.htm Please let us know if you and your colleagues are interested in this class to take advantage of group discount. Register now to secure your seat! Cheers, Elvis Miller, PhD Manager Training. XLSolutions Corporation 206 686 1578 [7]www.xlsolutions-corp.com [EMAIL PROTECTED] 2 Courses - (1) R/Splus Advanced Programming (2) R/Splus Fundamentals References 1. http://www.xlsolutions-corp.com/ 2. http://www.xlsolutions-corp.com/Radv.htm 3. http://www.xlsolutions-corp.com/Rfund.htm 4. mailto:[EMAIL PROTECTED] 5. mailto:[EMAIL PROTECTED] 6. http://www.xlsolutions-corp.com/training.htm 7. http://www.xlsolutions-corp.com/ 8. mailto:[EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
Jan T. Kim wrote: That's an idea I like very much too -- much better than the currently popular idea of protecting users from the unfriendliness of programming, anyway... It is just my opinion that the amount of mail in R-help speaks volumes about the current friendliness [1], or lack thereof, of R. Perhaps I am just the only one who thinks this way... [1] http://en.wikipedia.org/wiki/Usability __ LLama Gratis a cualquier PC del Mundo. Llamadas a fijos y móviles desde 1 céntimo por minuto. http://es.voice.yahoo.com __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] command completion?
[Duncan Murdoch] [Robert Citek] Does R have command or object name completion? [...] I don't think it would be a welcome change to the console versions; some of them use readline's filename completion which would almost certainly be broken by this. We have to put things in perspective, here. In my opinion, object name completion would be a lot more useful than filename completion, because in R, we name R objects much more often than we name files. Others need to run under ESS. While this is a good things for Emacs lovers, the requirement is rather unwelcome for pagans! :-) -- François Pinard http://pinard.progiciels-bpi.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] command completion?
Is the following a fortunes package candidate? Others need to run under ESS. While this is a good things for Emacs lovers, the requirement is rather unwelcome for pagans! :-) -- François Pinard http://pinard.progiciels-bpi.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html :-) -- Bert __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
When talking about user friendlyness of computer software I like the analogy of cars vs. busses: Busses are very easy to use, you just need to know which bus to get on, where to get on, and where to get off (and you need to pay your fare). Cars on the other hand require much more work, you need to have some type of map or directions (even if the map is in your head), you need to put gas in every now and then, you need to know the rules of the road (have some type of drivers licence). The big advantage of the car is that it can take you a bunch of places that the bus does not go and it is quicker for some trips that would require transfering between busses. Using this analogy programs like SPSS are busses, easy to use for the standard things, but very frustrating if you want to do something that is not already preprogrammed. R is a 4-wheel drive SUV (though environmentally friendly) with a bike on the back, a kayak on top, good walking and running shoes in the pasenger seat, and mountain climbing and spelunking gear in the back. R can take you anywhere you want to go if you take time to leard how to use the equipment, but that is going to take longer than learning where the bus stops are in SPSS. Now there are tools like Rcmdr that help get you started (maybe a gps unit in the R suv above), but if we make R too user friendly then we limit what can be done with it. I think the volume of mail in R-help is partly due to lack of friendliness, but a lot of it is due to the flexibility as well, if it did less, there would be less to learn and ask questions about. I for one prefer to do a little more work learning the program in exchange for being able to do a lot more with it. To mangle a famous Einstien quote: Statistical packages should be made as user friendly as possible, but no friendlier. -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Manuel López-Ibáñez Sent: Wednesday, May 10, 2006 2:19 PM To: r-help@stat.math.ethz.ch Subject: Re: [R] Can't there be a cd command? Jan T. Kim wrote: That's an idea I like very much too -- much better than the currently popular idea of protecting users from the unfriendliness of programming, anyway... It is just my opinion that the amount of mail in R-help speaks volumes about the current friendliness [1], or lack thereof, of R. Perhaps I am just the only one who thinks this way... [1] http://en.wikipedia.org/wiki/Usability __ LLama Gratis a cualquier PC del Mundo. Llamadas a fijos y móviles desde 1 céntimo por minuto. http://es.voice.yahoo.com __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] command completion?
On Wed, 10 May 2006 13:26:32 -0700 Berton Gunter wrote: Is the following a fortunes package candidate? ...promoted to fortunes package member in the devel-version. thx, Z Others need to run under ESS. While this is a good things for Emacs lovers, the requirement is rather unwelcome for pagans! :-) -- François Pinard http://pinard.progiciels-bpi.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html :-) -- Bert __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
...another fortunes package candidate? I especially liked the sections beginning R is a 4 wheel drive SUV..., but a lot of it is great IMHO. Well said! Bestimmt! -- Bert -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Greg Snow Sent: Wednesday, May 10, 2006 1:37 PM To: r-help@stat.math.ethz.ch Subject: Re: [R] Can't there be a cd command? When talking about user friendlyness of computer software I like the analogy of cars vs. busses: Busses are very easy to use, you just need to know which bus to get on, where to get on, and where to get off (and you need to pay your fare). Cars on the other hand require much more work, you need to have some type of map or directions (even if the map is in your head), you need to put gas in every now and then, you need to know the rules of the road (have some type of drivers licence). The big advantage of the car is that it can take you a bunch of places that the bus does not go and it is quicker for some trips that would require transfering between busses. Using this analogy programs like SPSS are busses, easy to use for the standard things, but very frustrating if you want to do something that is not already preprogrammed. R is a 4-wheel drive SUV (though environmentally friendly) with a bike on the back, a kayak on top, good walking and running shoes in the pasenger seat, and mountain climbing and spelunking gear in the back. R can take you anywhere you want to go if you take time to leard how to use the equipment, but that is going to take longer than learning where the bus stops are in SPSS. Now there are tools like Rcmdr that help get you started (maybe a gps unit in the R suv above), but if we make R too user friendly then we limit what can be done with it. I think the volume of mail in R-help is partly due to lack of friendliness, but a lot of it is due to the flexibility as well, if it did less, there would be less to learn and ask questions about. I for one prefer to do a little more work learning the program in exchange for being able to do a lot more with it. To mangle a famous Einstien quote: Statistical packages should be made as user friendly as possible, but no friendlier. -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Manuel López-Ibáñez Sent: Wednesday, May 10, 2006 2:19 PM To: r-help@stat.math.ethz.ch Subject: Re: [R] Can't there be a cd command? Jan T. Kim wrote: That's an idea I like very much too -- much better than the currently popular idea of protecting users from the unfriendliness of programming, anyway... It is just my opinion that the amount of mail in R-help speaks volumes about the current friendliness [1], or lack thereof, of R. Perhaps I am just the only one who thinks this way... [1] http://en.wikipedia.org/wiki/Usability __ LLama Gratis a cualquier PC del Mundo. Llamadas a fijos y móviles desde 1 céntimo por minuto. http://es.voice.yahoo.com __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] help with writing output from two different arrays to two columns in an output file
Hi, I am very new to R and I have written the following block of code to generate a gamma distribution for variable x (which is an array) and a function y whose array values depend on the individual array values of x. The code is as follows: n=1000 x=rgamma(n,1.5,2) y=vector(numeric,n) for (i in 1:n){ y[i]=(2937/5*exp(-1/1000*x[i])/x[i]) } now I want to know how I can use write/write.data or anything similar to write the output from x[i] and y[i] to two different colums simultaneously in an output file (say output.txt). Thanks in advance, Arnav [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Unique?
Hello, I have sample data set that looks like: YEARMONTH DAY CONTINUESPL TIMEFISH TIMEUNITAREACOUNTY DEPTH DEPUNIT GEARTRIPID CONVUNIT 19921 26 1 SP0073928 8 H 7 25 4 NA 100 02163399054 161 19921 26 1 SP0073928 8 H 7 25 4 NA 100 02163399054 8 19921 26 2 SP0004228 8 H 7 25 4 NA 100 02163399054 161 19921 26 2 SP0004228 8 H 7 25 4 NA 100 02163399054 8 19921 25 NA SP0052652 8 H 7 25 4 NA 100 02163399057 85 19921 26 NA SP0037940 8 H 7 25 4 NA 100 02163399058 70 19921 27 NA SP0072357 8 H 7 25 4 NA 100 02163399059 15 19921 27 NA SP0072357 8 H 7 25 4 NA 100 02163399059 20 19921 27 NA SP0026324 8 H 7 25 4 NA 100 02163399060 8 19921 28 1 SP0072357 8 H 7 25 4 NA 100 02163399062 200 How can I use unique to extract the rows that have repeated tripid's only, not a unique value for each variable but only for TRIPID. I then want to condense the unique values by summing the CONVUNIT for each unique value of TRIPID. I posted a similar question last week and received a sufficient answer of how to do this without using uniqe. The solution below worked just fine on this sample data set but the full data set has 446,000 rows of data and my computer and R simply cannot handle this follwing code on data this large. conds-by(Step4,Step4$TRIPID,function(x) replace(x[1,],CONVUNIT,sum(x$CONVUNIT))) Step5-do.call(rbind,conds) Thank you, Cameron Guenther, Ph.D. Associate Research Scientist FWC/FWRI, Marine Fisheries Research 100 8th Avenue S.E. St. Petersburg, FL 33701 (727)896-8626 Ext. 4305 [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On Wed, 10 May 2006 13:51:06 -0700 Berton Gunter wrote: ...another fortunes package candidate? Hehe, yes, this time I was quicker ;-) I especially liked the sections beginning R is a 4 wheel drive SUV..., but a lot of it is great IMHO. Well said! Indeed. Bestimmt! Ja, das auch... Z __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Splines?
I think I have a spline problem, and I would like to implement the solution in S. There are a lot of spline algorithms and I am looking for some direction on what is most appropriate. I need a spline that can be made to extrapolate certain data points as described below. Of course it needs to happen algorithmically as I have lots of this data. Can a Bezier be fit to data? Consider a material volume sampled in discrete non-uniform intervals. There is a continuous trend in properties that is integrated over each interval. We are aware of the specific form of the continuous trend, but it is concealed by the interval nature of the data. The specific need is to identify the 'true' location and value of the minima and maxima of the function representing the data. Perhaps an example... In the following figure, the numbers represent the sampled layers: the height of the bars is the value of the sampled property. The x axis is depth into the material volume. The asterisk represents the true value and location of the minima and maxima. We know that they are lower and higher than the interval sampled data, and (critically) that the value of the maxima or maxima is offset from the center of the sampled volume, depending on the trajectory of the change between sampled layers. Perhaps you can imagine a smooth line connecting the asterisks which preserves the area of the corresponding bars. * 3 3 *3 * 3 444 222 3 444 5*5 2*2 3 444 555 6*6 222 3 444 555 666 [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Unique?
On May 10, 2006, at 4:02 PM, Guenther, Cameron wrote: How can I use unique to extract the rows that have repeated tripid's only, not a unique value for each variable but only for TRIPID. I then want to condense the unique values by summing the CONVUNIT for each unique value of TRIPID. Thanks, Cameron, for this question. This type of manipulation would be relatively simple to do in a RDBMS (e.g. MySQL, PostgreSQL, Oracle, etc.) But I'm curious to see how one would do the same in R. So, if folks send you solutions off-list, please do post them back to the list. Regards, - Robert http://www.cwelug.org/downloads Help others get OpenSource software. Distribute FLOSS for Windows, Linux, *BSD, and MacOS X with BitTorrent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] help with writing output from two different arrays to two columns in an output file
Hello Arnav, On May 10, 2006, at 3:56 PM, Arnav Ghosh wrote: The code is as follows: n=1000 x=rgamma(n,1.5,2) y=vector(numeric,n) for (i in 1:n){ y[i]=(2937/5*exp(-1/1000*x[i])/x[i]) } now I want to know how I can use write/write.data or anything similar to write the output from x[i] and y[i] to two different colums simultaneously in an output file (say output.txt). One way would be to create a data frame. I tend to think of data.frames like sheets in a spreadsheet or tables in a database. Once you have your data in a data.frame, then you could write to/read from a file. For example, within R: # create the data.frame myDataFrame - data.frame(x,y) # write it to a file write.table(myDataFrame, file = output.txt) # read it from a file into a new data.frame newDataFrame - read.table(output.txt) BTW, one new resource I discovered was the wiki. The page you may be interested in: http://wiki.r-project.org/rwiki/doku.php?id=tips:data-io:import_table Regards, - Robert http://www.cwelug.org/downloads Help others get OpenSource software. Distribute FLOSS for Windows, Linux, *BSD, and MacOS X with BitTorrent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Keeping scientific format on assignment
I am new to R and can't seem to find and answer for this after search CRAN. I am calculating Box-Ljung stats over the lags 1-25 to build a table from the acf and pacf statistics. I want to retrieve the p-value from the box.test data frame and assign it in a vector. The problem is with the assignment that truncates the p-value to 0. For example, the box.test object has a p-value of 2e-14 when I do a-box.test.object$p-value; a; the value of a is 0 not 2e-14. How do I keep the precision and format of the p-value. I have tried as.numeric(...), changing options (digits and sciprec), and can't seem to figure it out. Thank you Joe __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Concatenating data frame
Hello, I have searched through the R-help archive and find that the easiest way to concatenate data records in a dataframe is to use rbind() I know we can do that using rbind, but it is slow when we are doing rbind thousands of times to a growing list, each time adding one or two records to the ever growing existing data because in existingRecords-rbind(existingRecords, aNewRecordToBeAdded), I am making a copy of the data each time rbind is called! Is there a way to avoid these data copying? Thank you in advance! Charles __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Legend titles in log plots broken? (ver. 2.2.1)
Legend titles work in linear plots: curve(1/x, xlim = c(0, 1)) legend(x = 'topright', inset = 0.04, legend = '1/x', lty = 1, title = 'Legend Title') But when you change to a log plot on either dimension things get screwy: curve(1/x, xlim = c(0, 1), log = 'y') legend(x = 'topright', inset = 0.04, legend = '1/x', lty = 1, title = 'Legend Title') If you save the value legend() returns you can look at it and see that it's messed up: l - legend(x = 'topright', inset = 0.04, legend = '1/x', lty = 1, title = 'Legend Title') l $rect $rect$w [1] 0.2349272 $rect$h [1] 0.2727899 $rect$left [1] 0.7618728 $rect$top [1] 1.9936 $text $text$x [1] 0.9188374 $text$y [1] 1.81174 R.Version() $platform [1] i686-redhat-linux-gnu $arch [1] i686 $os [1] linux-gnu $system [1] i686, linux-gnu $status [1] $major [1] 2 $minor [1] 2.1 $year [1] 2005 $month [1] 12 $day [1] 20 $svn rev [1] 36812 $language [1] R __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Unique?
If you only care about the sum of CONVUNIT by each TRIPID then you can use tapply i.e.: step4-data.frame(TRIPID=rep(c(111,222,333),3),CONVUNIT=rpois(9,40)) result-tapply(step4$CONVUNIT,INDEX=step4$TRIPID,FUN=sum) result 111 222 333 115 107 123 Is this what you wanted to do? I can't think of anything faster than tapply for your problem. I hope this helps Francisco From: Guenther, Cameron [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Subject: [R] Unique? Date: Wed, 10 May 2006 17:02:33 -0400 Hello, I have sample data set that looks like: YEAR MONTH DAY CONTINUESPL TIMEFISH TIMEUNIT AREACOUNTY DEPTH DEPUNIT GEARTRIPID CONVUNIT 1992 1 26 1 SP0073928 8 H 7 25 4 NA 100 02163399054161 1992 1 26 1 SP0073928 8 H 7 25 4 NA 100 021633990548 1992 1 26 2 SP0004228 8 H 7 25 4 NA 100 02163399054161 1992 1 26 2 SP0004228 8 H 7 25 4 NA 100 021633990548 1992 1 25 NA SP0052652 8 H 7 25 4 NA 100 0216339905785 1992 1 26 NA SP0037940 8 H 7 25 4 NA 100 0216339905870 1992 1 27 NA SP0072357 8 H 7 25 4 NA 100 0216339905915 1992 1 27 NA SP0072357 8 H 7 25 4 NA 100 0216339905920 1992 1 27 NA SP0026324 8 H 7 25 4 NA 100 021633990608 1992 1 28 1 SP0072357 8 H 7 25 4 NA 100 02163399062200 How can I use unique to extract the rows that have repeated tripid's only, not a unique value for each variable but only for TRIPID. I then want to condense the unique values by summing the CONVUNIT for each unique value of TRIPID. I posted a similar question last week and received a sufficient answer of how to do this without using uniqe. The solution below worked just fine on this sample data set but the full data set has 446,000 rows of data and my computer and R simply cannot handle this follwing code on data this large. conds-by(Step4,Step4$TRIPID,function(x) replace(x[1,],CONVUNIT,sum(x$CONVUNIT))) Step5-do.call(rbind,conds) Thank you, Cameron Guenther, Ph.D. Associate Research Scientist FWC/FWRI, Marine Fisheries Research 100 8th Avenue S.E. St. Petersburg, FL 33701 (727)896-8626 Ext. 4305 [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Unique?
Dear Cameron, This is not with unique, but it gets the job done. Just create a new variable that is the three variables concatenated together. Then, you can just sum by this variable, like the following: mymat - matrix(letters, ncol=3, nrow=260) mymat - as.data.frame(mymat) mymat$dat - rnorm(260) mymat$id - paste(mymat[,1], mymat[,2], mymat[,3]) aggregate(mymat$dat, list(mymat$id), sum) HTH, Dave. On 5/10/06, Robert Citek [EMAIL PROTECTED] wrote: On May 10, 2006, at 4:02 PM, Guenther, Cameron wrote: How can I use unique to extract the rows that have repeated tripid's only, not a unique value for each variable but only for TRIPID. I then want to condense the unique values by summing the CONVUNIT for each unique value of TRIPID. Thanks, Cameron, for this question. This type of manipulation would be relatively simple to do in a RDBMS (e.g. MySQL, PostgreSQL, Oracle, etc.) But I'm curious to see how one would do the same in R. So, if folks send you solutions off-list, please do post them back to the list. Regards, - Robert http://www.cwelug.org/downloads Help others get OpenSource software. Distribute FLOSS for Windows, Linux, *BSD, and MacOS X with BitTorrent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Dave Armstrong University of Maryland Dept of Government and Politics 3140 Tydings Hall College Park, MD 20742 Office: 2103L Cole Field House Phone: 301-405-9735 e-mail: [EMAIL PROTECTED] web: www.davearmstrong-ps.com Facts are meaningless. You can use facts to prove anything that's even remotely true. - Homer Simpson To this day, philosophers suffer from Plato's disease: the assumption that reality fundamentally consists of abstract essences best described by words or geometry. (In truth, reality is largely a probabilistic affair best described by statistics) - Steve Sailer The Unexpected Uselessness of Philosophy [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Concatenating data frame
Put the intermediate results in a list and then use do.call: result - list() for (i in 1:100){ result[[i]] - data.frame(id=sample(letters,1), value=i) } newDataFrame - do.call('rbind', result) On 5/10/06, Charles Cheung [EMAIL PROTECTED] wrote: Hello, I have searched through the R-help archive and find that the easiest way to concatenate data records in a dataframe is to use rbind() I know we can do that using rbind, but it is slow when we are doing rbind thousands of times to a growing list, each time adding one or two records to the ever growing existing data because in existingRecords-rbind(existingRecords, aNewRecordToBeAdded), I am making a copy of the data each time rbind is called! Is there a way to avoid these data copying? Thank you in advance! Charles __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Jim Holtman Cincinnati, OH +1 513 646 9390 (Cell) +1 513 247 0281 (Home) What the problem you are trying to solve? [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] post hoc comparison in repeated measure
I haven't heard from anyone with my previous post. I guess I should post my dataset and my code here: The dataset has one factor as treatment with 4 levels (treatment1, treatment2, treatment3 and control), and another factor as time (36 time points). On average Each treatment group has 10 subjects with each followed up at each time points. The response variable is numeric, serum protein amount. So the between subject factor is treatment, and the within subject factor is time. I ran a 2-way ANOVA without interaction with repeated measures considering time as the within subject factor: dat-read.table(dat.txt, sep='\t', header=T, row.names=1) attach(dat) ## run aov: fit-aov(x~as.factor(treatment)+as.factor(Time)+Error(as.factor(animal))) summary(fit) ## it should produce a ANOVA table as follow: Error: as.factor(animal) Df Sum Sq Mean Sq F value Pr(F) as.factor(treatment) 3 1405900 468633 3.0584 0.04051 * Residuals36 5516274 153230 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Error: Within Df Sum Sq Mean Sq F value Pr(F) as.factor(Time) 35 7298866 208539 22.796 2.2e-16 *** Residuals 1365 12487181 9148 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 As you can see, the treatnment effect is marginally significant. So I decided to do all pairwise comparisons on treatments at each time point using t tests. I tried 2 ways to do the t tests: one is the regular t tests, i.e. only using the data in comparison at each time point for t test computation, then yes, I do see a lot of significant comparisons; the 2nd approach is to use the MSE from ANOVA (i.e. 153230) as the error term for t test calculation; then I don't see any significant comparisons which I think is inconsistent with the marginally significant p value (0.04) for treatment main effect. Below is my code, did I do something wrong here, or my assumption above is not right: significant main effect doesnot necessarily mean at least one pairwise comparisons must be significant. ## to do pairwise regular t test: pval-list() for (j in seq(5,180,5) ) { pval[[as.character(j)]]-pairwise.t.test(dat2$x[dat2$Time==j],dat2$treatment[dat2$Time==j],pool.sd=F,var.equal=T,p.adjust.method='none')$p.value } sort(unlist(pval)[!is.na(unlist(pval))]) ## as you can see, there are indeed many significant pairwise comparisons ## to do pairwise t tests using MSE from ANOVA pval - NULL for (i in 1:3) { for (k in (i+1):4) { for (j in seq(5,180,5) ) { ## ANOVA post test pval - c(pval, 2*(1-pt(abs(mean(x[Time==jtreatment==k])-mean(x[Time==jtreatment==i]))/sqrt(153230*(1/sum(treatment[Time==j]==i)+1/sum(treatment[Time==j]==k))), df=sum(treatment[Time==j]==i)+sum(treatment[Time==j]==k)-2))) } } } sort(pval) ## As you can see, none of the p values is significant Can anyone explain? Thanks BTW, is there a R function that can do post hoc comparison on repeated measure ANOVA (from avo() with Error term)? __ x Timeanimal treatment 1 498 5 1 1 2 411 10 1 1 3 317 15 1 1 4 276 20 1 1 5 256 25 1 1 6 289 30 1 1 7 149 35 1 1 8 261 40 1 1 9 224 45 1 1 10 178 50 1 1 11 199 55 1 1 12 211 60 1 1 13 334 65 1 1 14 286 70 1 1 15 199 75 1 1 16 323 80 1 1 17 125 85 1 1 18 77 90 1 1 19 236 95 1 1 20 191 100 1 1 21 72 105 1 1 22 192 110 1 1 23 317 115 1 1 24 288 120 1 1 25 148 125 1 1 26 104 130 1 1 27 59 135 1 1 28 68 140 1 1 29 181 145 1 1 30 177 150 1 1 31 142 155 1 1 32 251 160 1 1 33 110 165 1 1 34 156 170 1 1 35 112 175 1 1 36 103 180 1 1 37 509 5 2 1 38 371 10 2 1 39 324 15 2 1 40 321 20 2 1 41 353 25 2 1 42 392 30 2 1 43 282 35 2 1 44 256 40 2 1 45 321 45 2 1 46 250 50 2 1 47 371 55 2 1 48 419 60 2 1 49 361 65 2 1 50 359 70 2 1 51 411 75 2 1 52 275 80 2 1 53 231 85 2 1 54
[R] lattice package plots
I am using the lattice packge for its levelplot and contourplot. Is it possible to adjust the line thickness of the 'box' and tickmarks in these plots? Thanks for the attention, Matt Sundling __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Concatenating data frame
And adding to Jim's solution, you may be able to further improve the speed of your code by pre-allocating the list size i.e result - vector(list,100) for (i in 1:100){ result[[i]] - data.frame(id=sample(letters,1), value=i) } newDataFrame - do.call('rbind', result) Cheers Francisco From: jim holtman [EMAIL PROTECTED] To: Charles Cheung [EMAIL PROTECTED] CC: r-help@stat.math.ethz.ch Subject: Re: [R] Concatenating data frame Date: Wed, 10 May 2006 19:02:15 -0400 Put the intermediate results in a list and then use do.call: result - list() for (i in 1:100){ result[[i]] - data.frame(id=sample(letters,1), value=i) } newDataFrame - do.call('rbind', result) On 5/10/06, Charles Cheung [EMAIL PROTECTED] wrote: Hello, I have searched through the R-help archive and find that the easiest way to concatenate data records in a dataframe is to use rbind() I know we can do that using rbind, but it is slow when we are doing rbind thousands of times to a growing list, each time adding one or two records to the ever growing existing data because in existingRecords-rbind(existingRecords, aNewRecordToBeAdded), I am making a copy of the data each time rbind is called! Is there a way to avoid these data copying? Thank you in advance! Charles __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Jim Holtman Cincinnati, OH +1 513 646 9390 (Cell) +1 513 247 0281 (Home) What the problem you are trying to solve? [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Contour plot overlayed with line plot
I apologize for what may be a newbie question: I have two sets of data, one is X,Y,Z data that I'd like to make a contour plot of (Z defining the contours), and a second set of X,Y data (Y as a function of X) which I would like made into a line plot OVERLAYED on the contour plot (X and Y from both plots are in the same units and will have overlapping values). Any suggestions on how to do this in R? Thanks! --j -- Jonathan A. Greenberg, PhD NRC Research Associate NASA Ames Research Center MS 242-4 Moffett Field, CA 94035-1000 650-604-5896 AIM: jgrn307 MSN: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] nlme plot residuals per group
Your example is not self contained, and I don't know enough to replicate it myself, so I can't tell you what caused it or how to fix it. However, I can outline the type of thing I've often done with this kind of problem: 1. First, can you get the plots you want using examples from the book and distributed with the 'nlme' package? If no, might that provide a simple, self-contained example for a refined post? 2. If you get the plots you want from one of the standard examples, how does your example that doesn't work differ from the published example that does? That comparison might provide the insight you need to solve the problem. If it doesn't, I would then experiment with both my example and the published example until I either solved my problem or produced a revision of the published example that illustrated my problem. That revised example might then provide the core for a refined post. 3. However, if it were me, I'd carry it further, making a local copy of the appropriate function, using 'debug', and walking through the code line by line until I found where it broke. By the time I've done this, I've typically figured out the problem. To do this, you need to know about the 'method' function, plus possibly 'getAnywhere'. hope this helps, Spencer Graves Osman Al-Radi wrote: dear list: I used the nlme library according to the great Pinheiro/Bates book, on R2.3, WinXp Lac.lme is an lme object with unbalanced data, group is a factor variable with three levels, when I tried to plot the residuals by group I got this error msg: plot(Lac.lme,resid(.,type='p')~fitted(.)|group) Error in limits.and.aspect(prepanel.default.xyplot, prepanel = prepanel, : need at least one panel Also When I try to use the auPred() function I get the follwoing error msg: plot(augPred(Lac.lme)) Error in tapply(as.character(object[[nm]]), groups, FUN[[dClass]]) : arguments must have same length Any suggestions? Thanks -- Osman O. Al-Radi, MD, MSc, FRCSC Fellow, Cardiovascular Surgery The Hospital for Sick Children University of Toronto, Canada __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Can't there be a cd command?
On May 10, 2006, at 3:37 PM, Greg Snow wrote: R is a 4-wheel drive SUV (though environmentally friendly) with a bike on the back, a kayak on top, good walking and running shoes in the pasenger seat, and mountain climbing and spelunking gear in the back. But mapping R to a SUV makes R sound too much like a car, which many people know how to drive. That is, someone who knows how to drive a car pretty much knows how to drive an SUV. The SUV analogy implies that someone who knows another computer languages, e.g C, perl, VB, Java, SQL, would be able to use R. Unfortunately, that is in all likelihood not the case. R has a rather specialized syntax it uses to manipulate data. It's not like C or perl or VB or Java or SQL. So, knowing those languages helps a little but not much. Because of that, I would say R is more like a helicopter, a HUEY perhaps. Regards, - Robert http://www.cwelug.org/downloads Help others get OpenSource software. Distribute FLOSS for Windows, Linux, *BSD, and MacOS X with BitTorrent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] post hoc comparison in repeated measure
This is how I would do it. arraychip - read.table(arraychip.dat, sep='\t', header=T, row.names=1) for (i in 2:4) arraychip[[i]] - factor(arraychip[[i]]) ## run aov: fit - aov(x ~ treatment + Time + Error(animal), data=arraychip) summary(fit) ## single stratum for the same ANOVA fit2 - aov(x ~ treatment + animal + Time, data=arraychip) summary(fit2) trt.means - model.tables(fit2, cterms=treatment, type=means) treat.means - as.vector(trt.means$tables$treatment) treat.n - as.vector(trt.means$n$treatment) names(treat.means) - levels(arraychip$treatment) ## this expression works in R and S-Plus mi.mj - outer(treat.means, treat.means, -) s2 - summary(fit2)$Mean Sq[2] ## S-Plus s2 - summary(fit2)[[1]][animal,Mean Sq] ## R s2.n - s2 / treat.n si.sj - sqrt(outer(s2.n, s2.n, +)) q.tukey - qtukey(.95, 4 ,36) /sqrt(2) mi.mj lower - mi.mj - q.tukey*si.sj upper - mi.mj + q.tukey*si.sj lower upper ## in S-Plus you can also use multicomp.default(treat.means, df.residual=summary(fit2)$Df[2], vmat=diag( ## rep( summary(fit2)$Mean Sq[2] ## , length(trt.means)) / treat.n ), method=tukey) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] large data set, error: cannot allocate vector
Robert, You are hitting a known problem with data.frame ... row names. The row names alone are the reason the data.frame takes 10 times more memory than the vector. Your 100MM integer vector takes 381MB when you scan() it right? [4bytes*10^8/1024^2] But when you try and create a data.frame instead, something like 3.8GB would be required by the data.frame. This is beyond the practical 32bit limit, and nothing you do with memory options will solve that. The good news is that Prof Ripley has fixed the problem with data.frame row names in the latest development version of R. You could try that, it should be much more efficient i.e. a data.frame with a single integer column length 100MM should have object.size 381MB, just like a vector. However, how many columns do you have to deal with?3GB allows 100,000,000 x 7 columns of integer in memory. That doesn't leave any room for copies, or types greater in size than integer, so you are still limited. Above that, as others have suggested, you need to connect to an RDMS, or go 64-bit is much easier if that is possible for you. I'd be interested to hear how you get on. Regards, Mark Message: 41 Date: Tue, 9 May 2006 15:27:58 -0500 From: Robert Citek [EMAIL PROTECTED] Subject: Re: [R] large data set, error: cannot allocate vector To: r-help@stat.math.ethz.ch Message-ID: [EMAIL PROTECTED] Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed On May 9, 2006, at 1:32 PM, Jason Barnhart wrote: 1) So the original problem remains unsolved? The question was answered but the problem remains unsolved. The question was, why am I getting an error cannot allocate vector when reading in a 100 MM integer list. The answer appears to be: 1) R loads the entire data set into RAM 2) on a 32-bit system R max'es out at 3 GB 3) loading 100 MM integer entries into a data.frame requires more than 3 GB of RAM (5-10 GB based on projections from 10 MM entries) So, the new question is, how does one work around such limits? [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] zero-inflated mixed models
I've never heard of a Hurdle model before, but have you looked at 'lmer' associated with library(lme4)? For documentation, see the following: * Douglas Bates. Fitting linear mixed models in R. R News, 5(1):27-30, May 2005 (www.r-project.org - newsletter). * vignette(MlmSoftRev) in library(mlmRev). (If you don't know about vignettes, you need to know that you get both an Adobe Acrobat file as well as an R script file to play with as you read the *.PDF. The *.PDF should open, at least under Windows, from 'vignette(MlmSoftRev)', and the *.R file should appear with 'edit(vignette(MlmSoftRev))' -- or 'Stangle(vignette(MlmSoftRev)$file)' if you use ESS.) hope this helps, Spencer Graves Jeff Miller wrote: Does anyone know of an existing R package or code to run a mixed Hurdle model? I found glmmADMB, but that seems to be ZIP. Any recommendations? Thanks, Jeff [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] ape comparative analysis query
Chris Knight wrote: This raises various questions: 1) Was I misleading myself that my independent contrasts were valid in the first place? I think partly. Independent contrasts were designed for data for tip taxa only, not ancestral state data. One of the steps in the algorithm requires a correction to account for the fact that ordinarily the values for traits at nodes are estimated and not known, basically lengthening the branch length to that node by a certain amount (See Felsenstein's original paper). If you have known ancestral states, then you should not make this correction. 2) What is it, if anything, about the root taxon that causes this issue, given that other taxa also have zero branch lengths? When converting a tree to a correlation matrix for use in GLS or GEE, the Brownian motion assumption implies that the covariances among taxa are represented by their shared branch length from the root, and the height of each terminal taxon is equal to the variance of the trait for that taxon. If the tree is ultrametric, then the variances are equal. In your case, the root has zero covariance with each taxon, and zero variance. Now, the error in the gls call is caused by a division by zero in the calculation of the correlation matrix, because the root has zero variance. try: vcv.phylo(tree) vcv.phylo(tree, cor=TRUE) 3) Is there any way of getting around this and including data on the root taxon, or am I better off just dropping it (ultimately I want to work with much larger trees (up to tens of thousands of taxa) where that one piece of information will become relatively less important) I'm surprised that you have known values for the root data. Are you sure that the root taxon is not actually an outgroup? If so, it will have some branch length (variance), and the model should fit OK. HTH, Simon. Any help very much appreciated, Chris I'm working with ape 1.8-2 in R 2.1.1 under ubuntu 'Breezy' linux (unfortunately 2.1.1 is the latest easily available in breezy) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] command completion? [Broadcast]
From: François Pinard [Duncan Murdoch] [Robert Citek] Does R have command or object name completion? [...] I don't think it would be a welcome change to the console versions; some of them use readline's filename completion which would almost certainly be broken by this. We have to put things in perspective, here. In my opinion, object name completion would be a lot more useful than filename completion, because in R, we name R objects much more often than we name files. Others need to run under ESS. While this is a good things for Emacs lovers, the requirement is rather unwelcome for pagans! :-) If ESS is too Catholic for you, JGR might be the Reformation... Andy -- François Pinard http://pinard.progiciels-bpi.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html