Re: [R] Problem with sas.get function in Hmisc

2006-12-15 Thread Jean Vidal
... >status <- system(paste(shQuote(sasprog), shQuote(sasin), "-log", >shQuote(log.file))) >... > > > -- > Hong Ooi > Senior Research Analyst, IAG Limited > 388 George St, Sydney NSW 2000 > +61 (2) 9292 1566 > -Original Message---

Re: [R] Problem with sas.get function in Hmisc

2006-12-13 Thread Jean Vidal
The workaround you point seems to be the solution for the moment. I not sure I will have time (and courage) to try to figure out where the bug comes from. Hope a more literate R programmer than me is willing to dive in... Thank you for your answer. 2006/12/12, Agustin Calatroni <[EMAIL PROTECTED]

[R] Problem with sas.get function in Hmisc

2006-12-11 Thread Jean Vidal
Thomas, As F Harrel in a preceding message told me that you are the maintainer of Hmisc package, I write directly to you, with copy to the Rhelp list in case someone encountered the same problem and find some benefit in the response. I used quite often the function sas.get in Hmisc library which

[R] Heatmap.2 in gplots

2006-11-13 Thread Jean Vidal
I used the function heatmap.2 some times ago, and it worked as intended (for me...). I tried to reuse my old R program and I encounter some trouble. I suppose that something was changed in newer versions. One of the examples given in the help page shows the same problem : -- this part

Apologies --- Re: [R] Question regarding mosaicplot

2005-03-10 Thread Jean Vidal
n Tue, 18 Nov 2003 23:32:05 Jean Vidal wrote: > I tried this : > > mosaicplot(stoc ~ q9r + segca,data=tmp2,color=T) : works fine. > > And now, this : > > mosaicplot(stoc ~ q9r + segca, data=tmp2, color=T, main="Big title") > Error in model.frame(formula

[R] Question regarding mosaicplot

2005-03-10 Thread Jean Vidal
I tried this : mosaicplot(stoc ~ q9r + segca,data=tmp2,color=T) : works fine. And now, this : mosaicplot(stoc ~ q9r + segca, data=tmp2, color=T, main="Big title") Error in model.frame(formula, rownames, variables, varnames, extras, extranames, : invalid variable type I'm probably stupid

[R] Problem with R version 2.0.0 (and patched)

2004-10-14 Thread Jean Vidal
What I am doing : create a simple correlation matrix on 41 variables, then plot an heatmap with this program : library(gplots) mat<-cor(temp.alln,use="pairwise.complete.obs") hm<-heatmap.2(mat,symm=T) HM<-format(round(mat[hm[[1]],hm[[2]]],2)) library(RColorBrewer) brewer.pal(10,"Spectral")->mp he

[R] Question regarding mosaicplot

2003-11-18 Thread Jean Vidal
I tried this : > mosaicplot(stoc ~ q9r + segca,data=tmp2,color=T) : works fine. And now, this : > mosaicplot(stoc ~ q9r + segca, data=tmp2, color=T, main="Big title") Error in model.frame(formula, rownames, variables, varnames, extras, extranames, : invalid variable type I'm probably stu

[R] Problem with sas.get function in Hmisc library

2003-09-28 Thread Jean Vidal
I tried to use for the first time the Frank Harell's sas.get function (in Hmisc library) and cannot manage to import my sas datas. When I use the sas.get(library,member, ...) method, I get the following error : sas.get("pme","Edfv1_a")->test.import.sas Error in sas.get("pme", "Edfv1_a") : librar

[R] Plot of Canonical Correlation Analysis

2003-03-29 Thread Jean Vidal
[EMAIL PROTECTED] wrote: > Dear all, > > I didn't find any graphical solution in the package "mva" to plot the > canonical scores from a CCA (canonical correlation analysis). > Does anybody knows how to plot or has anybody already programmed : > - the map of the canonical scores, > - the graph of t