Probit dose-response model with second, random main effect
in [SAS]
|
Prev: modeling time series data using Generalized Linear Model in
Next: Running SAS on PC, need to delete old compiled macro
From: John Uebersax on 8 Mar 2010 19:12 Can anyone suggest a way in SAS to fit a probit dose-response regression model with a second main effect, e.g., probit(response) = f(dose, lab, error) where lab is a random effect? Proc probit will allow the second main effect, but not, as far as I can see, as a random effect. There are so many procs (MIXED, NLMIXED, GLM...) I'm not sure where to look. Thanks in advance for any suggestions. John Uebersax
From: Oliver Kuss on 9 Mar 2010 02:59
On 9 Mrz., 01:12, John Uebersax <jsueber...(a)gmail.com> wrote: > Can anyone suggest a way in SAS to fit a probit dose-response > regression model with a second main effect, e.g., > > probit(response) = f(dose, lab, error) > > where lab is a random effect? > > Proc probit will allow the second main effect, but not, as far as I > can see, as a random effect. > > There are so many procs (MIXED, NLMIXED, GLM...) I'm not sure where to > look. > > Thanks in advance for any suggestions. > > John Uebersax Dear John, as your response is binary, you should use PROC GLIMMIX. It allows a probit link for the linear predictor and random lab effects. PROC NLMIXED would also be an option, however, you would have to code the probit likelihood yourself. PROC MIXED is only suited for continuous responses. Hope that helps, Oliver |