[S] singularity and lrm.fit

carlos (carlos@pcweb.net)
Wed, 11 Mar 1998 12:10:22 -0800


Is there a reason why the lrm function in the design library would be
more sensitive to singularity than the glm function? And is there a
work-around? I've tried reducing the tol argument, but with no success.
Upon reading the articles in the S-lib archive, I'm still unclear on why
this is happening. I'm using Splus4/WinNT.

flrm2_lrm(formula = OVRES == "F" ~ FOC10DEN + FOC10SIZ + FOC10CON +
PRECIPANN + UTMN + I(UTMN^2) + FOC10DEN:FOC10CON +
FOC10SIZ:PRECIPANN,data = fpoints5)
singular information matrix in lrm.fit (rank= 1 ). Offending
variable(s):
FOC10DEN FOC10SIZ FOC10CON PRECIPANN UTMN FOC10DEN * FOC10CON FOC10SIZ
* PRECIPANN
Error in 1:nrp: No data to interpret as numerical value
Dumped

-- 
Carlos Carroll, M.S.
Conservation Biology Institute and
Dept. of Forest Science, Oregon State University
Mailing address: PO Box 104
Orleans, CA 95556
e-mail: carlos@pcweb.net
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu.  To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message:  unsubscribe s-news