[R] Build a matrix using JRI

Hi everyone I'm a new user of R and JRI for a project (debian + eclipse + Java + JRI). I need to build a matrix from datas and i proceed like this : First step : create the matrix (Rengine re) re.idleEval(A - mat.or.vec(10,40),true); Second Step : feed the structure re.idleEval(A[1,1] -

Re: [R] A problem with copy() in the svIO package

Hi Kenneth, what did the package maintainer of the svIO answer to you when you asked him about this? I think the posting guide asks you pretty explicitly to ask the package maintainer *) about such problem. *) You find the maintainer e.g., from packageDescription(svIO) or library(help

Re: [R] Drawing a 3-D plot for PCA?

On Thu, 2006-12-21 at 16:28 -0600, Frank Duan wrote: Hi All, Can anyone point me a hint (package) how to draw a 3D plot using the first 3 components from PCA? Thanks a lot, FD See ?ordiplot3d and ?ordirgl in package vegan. rda() in that package can be used to perform PCA, which can

[R] Kalman Filter in Control situation.

I am looking for a Kalman filter that can handle a control input. I thought that l.SS was suitable however, I can't get it to work, and wonder if I am not using the right function. What I want is a Kalman filter that accepts exogenous inputs where the input is found using the algebraic

[R] Pb with R.matlab

Hello, I have problems running the R.matlab package. I work with R 2.3.1, under windows (using the Rgui). It seems that, even after loading the package, the simple functions such as writeMat() or readMat() are not recognized. For example, the following script leads to an error:

[R] ape-package

Dear Sir or Madam, I am very new to R, and I am trying to install seqinr-package. In the manual I read that I need to install ape-package first, and I failed to do it. I had the following error messges: /usr/bin/ld: cannot find -lgfortran collect2: ld returned 1 exit status make: *** [ape.so]

Re: [R] Pb with R.matlab

Hi you probably installed (downloaded and unpacked) a package but you forgot to load it to R and therefore R does not know that you want to use it. library(R.matlab) see ?library and R-intro.html HTH Petr On 22 Dec 2006 at 11:17, Arthur Leblois wrote: Date sent: Fri, 22 Dec

Re: [R] Pb with R.matlab

Thank you. This was a mistake from me, I actually did not load the package. Sorry for this! Arthur On Fri, 22 Dec 2006, Petr Pikal wrote: Hi you probably installed (downloaded and unpacked) a package but you forgot to load it to R and therefore R does not know that you want to use it.

Re: [R] Math expression with object to evaluate

Use bquote: ID - 3 plot(1,1, main = bquote(ID ~ is ~ .(ID))) Also please read the last line of every message to r-help regarding reproducible examples. (None of your variables were defined.) On 12/22/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hello. I have a question that probably has a

Re: [R] ape-package

On Fri, 2006-12-22 at 14:13 +0300, Natasha Sernova wrote: Dear Sir or Madam, I am very new to R, and I am trying to install seqinr-package. In the manual I read that I need to install ape-package first, and I failed to do it. I had the following error messges: /usr/bin/ld: cannot find

Re: [R] Pb with R.matlab

Hello Arthur, Load the package first before using functions: library(R.matlab) Regards, François -Message d'origine- De : Arthur Leblois [mailto:[EMAIL PROTECTED] Envoyé : vendredi 22 décembre 2006 11:17 À : r-help@stat.math.ethz.ch Objet : [R] Pb with R.matlab Hello, I have

Re: [R] newbie estimating survival curve w/ survfit for coxph

fit - coxph(Surv(futime,fustat)~ age +strata(rx), data=ovarian, subset=1:23) curves - survfit(fit, newdata=ovarian[24:26,]) I don't think this is mentioned in the documentation (I'll have to fix that!), but subscripting works for survfit objects. In this case there are

Re: [R] odfWeave problem

Hi, Doesn't work is kind of vague... what is or isnt' happening? But in this case, I think it might be easy: On 12/22/06, Abhijit Dasgupta [EMAIL PROTECTED] wrote: Hi, StyleDefs$ArialNormalRight$textAlign='end' I think this should be 'right' rather than 'end. The ODF XML specification is

Re: [R] odfWeave problem

Sorry, I zipped you off a reply without actually having OpenOffice available - never wise. First, both right and end work for me as alignment specifications, though OpenOffice itself seems to use end. Here's the R code I'm using to define the styles: # Now to specify the styles

Re: [R] lda plotting: labeling x axis and changing y-axis scale

--- Wade Wall [EMAIL PROTECTED] wrote: Hi all, I have performed an lda on two groups and have plotted using plot(x.lda), with x.lda being my lda results. I have forgotten how to change the labels of the of the x-axes (they are currently listed as Group1 and Group 13), and to rescale

[R] absent

Je serai absent(e) à partir du 22/12/2006 de retour le 27/12/2006. Je vous répondrais dès mon retour le 27/12. Si urgence, merci de contacter M.Bidart (5503). Si vous n'etes pas destinataires de ce message, merci d'aver...{{dropped}} __

Re: [R] Kalman Filter in Control situation.

you have to be really careful because There are two versions of the kalman filter in terms of notation ( even where you don't have exogenous ) Y_t = F_tprime*theta_t-1 + epsilon_t G_t = G_t*theta_t-1 + omega_t - Y_t = F_tprime*theta_t + epsilon_t G_t =

[R] question

I have a question about R. I would like to simulate a following model. Xt+1=Xt(1+b) b is a random variable. to get 10 plots. I haven't used R program, how can I code on R to run above simulation? It rould be grateful if you kindly answer... thank you. yumiko watanabe

[R] Pelora problem

Hi all, I'm using Pelora (supclust) package for supervised clustering of a microarray dataset. The original package does not delete the genes that are present in the earlier clusters for further cluster identification. This leaves us with mostly redundant clusters. I played around with the source

[R] plot.mids / Error in plot.new() : figure margins too large

Hello R-Users, I would like to check the convergence of my imputations. However, when I use the function the plot.mids(), I always obtain the following error message Error in plot.new() : figure margins too large I read the same question in thread from November 2005 (see below). I

Re: [R] iplots/JGR on OS X 10.4.8

Pointers only hopefully of some help. iplots should work as you have described, however there have been a number of different versions of JGR each for a different version of R - have you got the correct version? I believe JGR shouldn't work as you have described, even though I have no experience

Re: [R] question

You will not learn anything unless you do your own homework. Watanabe Yumiko [EMAIL PROTECTED] wrote: I have a question about R. I would like to simulate a following model. Xt+1=Xt(1+b) b is a random variable. to get 10 plots. I haven't used R program, how can I code

Re: [R] writing R extension

ahmad ajakh [EMAIL PROTECTED] wrote: Hi all, I am dealing with the same issue here and I was wondering whether it would be possible to just save the R compliled function objects in a directory and just attach the directory to the search path. (I am using R2.4.0+ESS+Xemacs in windows XP).

Re: [R] lda plotting: labeling x axis and changing y-axis scale

Sorry I wasn't clearer. I believe that it was a specialized function, but it may have been plot(). What I am basically trying to do is alter the y-axis to represent frequency and change the labels on the plotting of the linear discriminant analysis results. I can't seem to do this with plot(),

Re: [R] multinom(nnet) analogy for biglm package?

Dear Prof. Ripley, Many thanks for your reply, especially during the holiday season! From: Prof Brian Ripley [mailto:[EMAIL PROTECTED] You seem to believe a multinomial logistic regression is a GLM: it is not. With one line, you've struck the heart of one of my problems. As a follow-up

[R] unexpected conversion from POSIXlt to POSIXct

The documentation for DateTime classes in R 2.3.0 says that limited arithmetic is available for both POSIXlt and POSIXct. It needs to make clear that POSIXlt can be converted to POSIXct without warning under arithmetic operations EXAMPLE ISOdatetime(2003, 1, 1, 0, 0,0) - dd as.POSIXlt(dd) - dd

Re: [R] question

Sorry, but might as well not have answered..., no? Yumiko, you didnt say what your distribution assumptions were. take a look at ?runif and ?rnorm and use that to search for help on your specific topic to get you started. -Tariq On 12/22/06, Mike Prager [EMAIL PROTECTED] wrote: You will not

Re: [R] multiline system call

Look at the help page for ?system, especially the 'input=' arg. Maybe this sort of strategy will fly: system(sh,intern=T,input=c(echo x,echo y)) [1] x y In your case, maybe system('sh' input=c('sendmail -f [EMAIL PROTECTED] -t [EMAIL PROTECTED]','test','.') or system('sendmail -f

Re: [R] Kalman Filter in Control situation.

Todd Remund wrote: I am looking for a Kalman filter that can handle a control input. I thought that l.SS was suitable however, I can't get it to work, and wonder if I am not using the right function. What I want is a Kalman filter that accepts exogenous inputs where the input is found

[R] heatmap with levelplot?

Hi, How do I anchor z=0 to the white color in a levelplot so that the color changes from cyan to magenta precisely as z changes from negative to positive? Also is it easy to change color scheme, say to blue/red as it's more dramatic? Is there a better function for showing heatmap with a color

Re: [R] multiline system call

Dear Sir, Le vendredi 22 décembre 2006 20:05, vous avez écrit : system(sh,intern=T,input=c(echo x,echo y)) Thanks for the advice, but I do not find doc about the input arg for the system function. Besides, $ system(sh,intern=T,input=c(echo x,echo y)) Erreur dans system(sh, intern = T, input =

Re: [R] multiline system call

Its only available on Windows. On 12/22/06, Nicolas Mazziotta [EMAIL PROTECTED] wrote: Dear Sir, Le vendredi 22 décembre 2006 20:05, vous avez écrit : system(sh,intern=T,input=c(echo x,echo y)) Thanks for the advice, but I do not find doc about the input arg for the system function.

Re: [R] heatmap with levelplot?

On 12/22/06, Yuli Zhou [EMAIL PROTECTED] wrote: Hi, How do I anchor z=0 to the white color in a levelplot so that the color changes from cyan to magenta precisely as z changes from negative to positive? The changepoints are defined by 'at', and the colors are chosen more or less linearly, so

[R] building R-package under windows - error - in options(deafultPackages) was not found

Dear all, I have been building R packages under windows on my old pc, successfully. Now I have bought a new pc, still running windows, and I am trying to build the same R packages as before, but now without the same success. I have installed the Rtools, perl, mingw and added them to the

Re: [R] multiline system call

On Fri, 22 Dec 2006, Gabor Grothendieck wrote: Its only available on Windows. Right, but on linux this works system(echo x\necho y,intern=T) [1] x y system(echo x ; echo y,intern=T) [1] x y so you don't really need input=... But rereading Nicolas original post, it looks like the

Re: [R] writing R extension

Thanks Mike. AA. - Original Message From: Mike Prager [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Friday, December 22, 2006 12:45:40 PM Subject: Re: [R] writing R extension ahmad ajakh [EMAIL PROTECTED] wrote: Hi all, I am dealing with the same issue here and I was wondering

Re: [R] nonparametric significance test for one sample

Hi, Greg: Just let you know that the beta regression package in R can only work for data on the open interval (0, 1) Do you know any good test of mean for beta distribution? How to verify if the data is beta distributed? For example, I may want to test the null : mean = 0.01 I think

Re: [R] nonparametric significance test for one sample

HelponR suncertain at gmail.com writes: Hi, Greg: Just let you know that the beta regression package in R can only work for data on the open interval (0, 1) Do you know any good test of mean for beta distribution? How to verify if the data is beta distributed? For example, I may

[R] bug in odfWeave

Hi, I think there is a minor bug in odfWeave. In the function odfStyleGen, the following line has an extra =: if(length(grep(italic, thisStyle$fontType))) fontText - c(fontText, tagattr(fo:font-style=, italic)) This is creating an error if some text needs to be

Re: [R] nonparametric significance test for one sample

Seems there is an exact non-parametric Bayes test for mean of non-negative numbers as said on http://www.toad.net/~jkaplan2/martMean.htmhttp://www.toad.net/~jkaplan2/martMean.htm#From%20bins I will check if R may have such package. On 12/22/06, Ben Bolker [EMAIL PROTECTED] wrote: HelponR

[R] complex barplot enquiry

Hello, I was hoping for some advice to resolve a problem I am having trouble with. The data consists of a series of pre and post variables, in a dataframe called 'offend'. I am interested in graphically depicting the pre post values for a factor variable called 'decision' which has 4 values :

Re: [R] multiline system call

Le vendredi 22 décembre 2006 23:31, Charles C. Berry a écrit : Perhaps a workaround is to pipe the input into sendmail (I recall sendmail accepts input from sdtin) from 'echo' This works fine this way. Thanks a lot. (but note that my 2.4.0 linux binary does not accept the input=xyz as an

Re: [R] multiline system call

On Sat, 23 Dec 2006, Nicolas Mazziotta wrote: Le vendredi 22 décembre 2006 23:31, Charles C. Berry a écrit : Perhaps a workaround is to pipe the input into sendmail (I recall sendmail accepts input from sdtin) from 'echo' This works fine this way. Thanks a lot. (but note that my 2.4.0 linux