hein0106@UMN.EDU wrote: > >I am trying to analyze a data set of 300 observations. The response >variable is a 0 or 1 (binary). I also want to include a random effect in >the model. > >What would be the best option PROC GLIMMIX or NLMIXED?

What are your data like? What is the sample size, and what are your variances like, and how big are the individual cells here?

>The model that I would like to fit is below: > >CLASS BREED HY herd service cow; >MODEL CONCEIVE = HERD breed hy(herd) service; >random cow(breed); > > >I can not seem to get PROC GLIMMIX to converge with binary data and I am >not >sure where to get starting values to use NLMIXED?

Hmm.

What is your *exact* code for PROC GLIMMIX? I'm used to PROC GLIMMIX working pretty well for binary dependent variables.

Why do you have to get starting values for NLMIXED? (If you're desperate, try something simple like proc mixed for that.)

>If anyone could offer suggestions, I would appreciate it. > >Brad H.

HTH, David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330

_________________________________________________________________ Make every IM count. Download Messenger and join the i’m Initiative now. It’s free. http://im.live.com/messenger/im/home/?source=TAGHM_June07