Dear college, I'm trying to fit a multinomial logit model in SAS' CATMOD procedure, and I wish to model the logit ln[pr(A=a|covariates)/(1-pr(A=a|covariates))] for a=(1, ..., 8). However, I'm unable to find the command that specifies the logits in this way. The command response=logits leads to the modeling of ln[pr(A=a|covariates)/pr(A=8|covariates)] for a=(1, ..., 7) and this is not what I'm interested in. Does anyone know the proper response command? Thanks in advance, Jarl Kampen IPSoM (Institute for Political Sociology & Methodology), Faculty of Social & Political Sciences, Catholic University of Brussels, Vrijheidslaan 17, 1081 Brussels, Belgium. Voice +32(0)2 412 43 38. Fax +32(0)2 412 42 00. E-mail [log in to unmask] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%