[S] predict.lme

Regula Schwager-Suter (suter@inw.agrl.ethz.ch)
Thu, 12 Nov 1998 11:56:05 +0100


I am using S-Plus, Version 4.5 Release 1 on Windows-NT

I have a dataset with repeated measurements from 213 animals. I fitted an
lme-object with repeated measurements from 212 animals. The intercept is
the only random effect. The model contains severeal fixed effects, with the
contrast-option set to "contr.treatment" due to unbalancedness in the data
(as proposed in Venables & Ripley). Then I wanted to predict the y-value of
the 213th animal as follows:

> predizione_predict.lme(u[[1]], estim)
Error in contr.treatment("32", T): Contrasts not defined for 0 degrees of
freedom
Dumped

The data.frame "estim" contains 30 records of the 213th animal. The animal
has in each record the same fixed effect race (=32) and also the same fixed
effect lactation number (=2), which obvioulsy causes the "predict.lme"
function to dump.

Is there a possibility to avoid this problem? Or did I miss anything
concerning the "predict.lme" function?

Thank you in advance for your help

Best regards Regula
_____________________________________

Regula Schwager-Suter
Swiss Federal Institute of Technology
Institute of Animal Science
Animal Breeding
Clausiusstrasse 50
CH-8092 Zürich
phone: +41 (0)1 632 39 56
fax: +41 (0)1 632 12 60
e-mail: suter@inw.agrl.ethz.ch
http://www.tz.inw.agrl.ethz.ch/~suter
_____________________________________
-----------------------------------------------------------------------
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