s-news
[Top] [All Lists]

Re: problem with summary multinom

To: <s-news@lists.biostat.wustl.edu>
Subject: Re: problem with summary multinom
From: "Barker, Chris" <Chris.Barker@cvt.com>
Date: Fri, 16 Jan 2009 12:29:34 -0800
In-reply-to: <61088.89.54.16.155.1232137302.squirrel@mail.uni-bayreuth.de>
Thread-index: Acl4GCZXKoRwf3dvQim63xLzwrhAWgAANM0g
Thread-topic: [S] problem with summary multinom
 
It may  be helpful  to see your function call.


-----Original Message-----
From: s-news-owner@lists.biostat.wustl.edu
[mailto:s-news-owner@lists.biostat.wustl.edu] On Behalf Of Christian
Bauer
Sent: Friday, January 16, 2009 12:22
To: s-news@lists.biostat.wustl.edu
Subject: [S] problem with summary multinom

Hi all,

I understand that multinorm belongs to a package that belongs to a book.
Still I have a hopefully easy to answer question. I get the following
error message when I call the summary of a multinom regression:

svd.default(X): x must be numeric or complex

I do not understand the problem. Is it because the algorithm did not
converge? The regression ends with

iter 100 value 298.219000
final  value 298.219000
stopped after 100 iterations

The data set is fairly unbalanced with roughly 4000 times 0 and 30 times
1,2,3 each. I'd appreciate any comments.

I use Splus 7.0 Build B7487 on a fairly new PC and WindowsVista.

Best,
Christian






--------------------------------------------------------------------
This message was distributed by s-news@lists.biostat.wustl.edu.  To
unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
the BODY of the message:  unsubscribe s-news


<Prev in Thread] Current Thread [Next in Thread>