Hi @Cochetti, As I was explaining in my last reply, in this link there is an example of a model with class variables which still writes out the Odds Ratio result without any problem, that's why I thought it is not related to that. Let me paste the example code here: proc glimmix data=uti;
freq count;
class diagnosis treatment;
model response = diagnosis treatment diagnosis*treatment / dist=binary;
lsmeans diagnosis*treatment / slicediff=diagnosis oddsratio ilink;
run; My case is pretty similar except from the fact that I'm using the default dist=Gaussian (and its corresponding link=identity) because if I use dist=binary (whose default link=logit) my model is not converging. Let me remind you my original question. In SAS documentation it says (or at least that's what I understand) that you can only calculate the Odds Ratio through the LSMESTIMATE statement and EXP option for the LINK= LOGIT | GLOGIT | CLOGIT. Is it correct to assume then, if I'm using a different link function (link=identity), that the exponentiated estimate can still be interpreted as the Odds Ratio? Let me quote the SAS documentation page that I'm referring to (its link is here😞 In models with a logit, generalized logit, or cumulative logit link, you can obtain estimates of odds ratios through the ODDSRATIO options in the PROC GLIMMIX, LSMEANS, and MODEL statements. This section provides details about the computation and interpretation of the computed quantities. Note that for these link functions the EXP option in the ESTIMATE and LSMESTIMATE statements also produces odds or odds ratios. Thanks again for your help and patience.
... View more