[R] problem using MASS corresp and mca functions

[Alexandre Fayolle]

> > I'm an absolute beginner with R and neophite in data analysis, so
> > please bear  with me if I ask stupid question.
> > 
> > I'm trying to do a correspondence analysis using R and MASS corresp
> > function,  but I get an error message which I'm unable to interpret:
> > 
> > data(weblog)
> 
> Try 
>   read.table(filename, sep=";", header=TRUE)
> instead of data(weblog). I guess the file is not read in correctly (?).

I get another error with this method:
> corresp(~ url + fromurl, data=log)
Error in svd(t(t(x1 * Dr) * Dc)) : error  34  in dsvdc
(not that this is easier to fix...)

> > > library(MASS)
> > > corresp(~ url + fromurl, data=weblog)
> > Error in svd(t(t(x1 * Dr) * Dc)) : error  306  in dsvdc

> Works for me, so I think it is the way you read in the data. Have you
> checked, if the data.frame weblog looks OK in R?

It does, as far as I can tell. The file has no empty fields, only
qualitative variables... I tried to suppress some lines in the file, and
finally reached the point where deleting the last line would get the thing
running. So I thought Ahaa, there you are, however the line itself has
nothing special. And replacing it with another line from somewhere earlier
in the file would break things again. 

If anyone wants to help me, I can send the file (it's only 2.3K gzipped). 

> [Assuming you are running R-1.1.1 and the latest MASS package]
 
I am : I installed R on a Redhat 6.2 box with latest RPM from CRAN, and
MASS from VR_6.2-1.tar.gz


Alexandre Fayolle
-- 
http://www.logilab.com 
Narval is the first software agent available as free software (GPL).
LOGILAB, Paris (France).






-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._