[R] problem using MASS corresp and mca functions

Subject: [R] problem using MASS corresp and mca functions
From: Alexandre Fayolle (Alexandre.Fayolle@logilab.fr)
Date: Mon 13 Nov 2000 - 00:57:43 EST

• Next message: Uwe Ligges: "Re: [R] putting output from print() into a string?"
• Previous message: David Firth: "[R] putting output from print() into a string?"
• Maybe reply: Alexandre Fayolle: "[R] problem using MASS corresp and mca functions"

Message-ID: <Pine.LNX.4.21.0011121542200.14067-100000@leo.logilab.fr>

> > 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@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

• Next message: Uwe Ligges: "Re: [R] putting output from print() into a string?"
• Previous message: David Firth: "[R] putting output from print() into a string?"
• Maybe reply: Alexandre Fayolle: "[R] problem using MASS corresp and mca functions"

This archive was generated by hypermail 2b25 : Thu 01 Feb 2001 - 16:14:32 EST