See this note.  ... View more

The GENMOD analysis of the above example would be the following. No WEIGHT statement (or ZERO option) is needed. But with this modeling approach, one of the parameters - the one for the level with zero count - is infinite. For the level without the zero count, a confidence interval is provides but is wider than the intervals from PROC FREQ. The nonmodeling approach provided by PROC FREQ avoids parameter estimation and the exact analysis provides good estimates and confidence intervals for both levels.    data b; input t c; tr=20; off=log(tr); datalines; 1 0 0 10 ; proc genmod; class t; model c=t/d=p offset=off; lsmeans t / ilink cl; run; ... View more

You can get risk estimates and confidence intervals using the RISKDIFF option in PROC FREQ. In this example with two treatments, note that one treatment (t=1) has a zero count for the event (y=1). The "Column 2 Risk Estimates" table provides confidence intervals for the risk in each treatment. If you also want an exact confidence interval for the risk difference, you can add an EXACT statement with the RISKDIFF option.     data a; input y t c; datalines; 1 1 0 0 1 20 1 0 10 0 0 10 ; proc freq; table t*y / riskdiff; weight c / zeros; run;   ... View more

See this note.   Note that with a log-linked model, your offset variable should generally be the log of the duration, not the duration itself. ... View more

Note that the presence of the CLASS statement does not make the model a multinomial model. In PROC LOGISTIC, the model is a multinomial model if the response variable has more than two distinct values. Further, a multinomial model can be a ordinal or nominal multinomial model depending on whether the multiple levels of the response have a natural ordering or not. PROC LOGISTIC fits an ordinal model by default (using cumulative logits) when there is a multilevel response. If the response level are not ordered, then the LINK=GLOGIT option in the MODEL statement can be specified to fit a nominal multinomial model.    Now to your question. As with the binary logistic model, you can use the SCALE=NONE and AGGREGATE options in the MODEL statement to obtain Pearson and deviance statistics which are valid goodness of fit tests only if there is sufficient replication within the subpopulations defined by the distinct covariate patterns in the data. For purposes of comparing competing models you can use the R-square statistic available with the RSQUARE option in the MODEL statement. Also for comparing models, you could use the AIC and SC statistics which are provided by default. Of course, you can always save predicted probabilities from the model using the PREDPROBS=INDIVIDUAL option in the OUTPUT statement. Similarly, you can score a new set of data and save predicted probabilities using the SCORE statement.  See the example titled "Scoring Data Sets" in the LOGISTIC documentation.  ... View more

As you describe it, you essentially want to get predicted means for several specified settings of all of your predictors. Save your fitted model by adding a STORE statement in GLIMMIX. You could then simply create a data set containing an observation for each of the settings for which you want estimated means, and then use PROC PLM to read the stored model information and specify a SCORE statement to obtain the predicted values for the data set you created.  ... View more

If this is, as it seems, a finite mixture model, then you might find it easier to implement in PROC FMM. ... View more

What you want to do can be done using the DIFF=ANOM option in the LSMEANS statement in PROC LOGISTIC.  See this note. ... View more

The zero divide problem only exists in SAS 9.4 TS1M2 and earlier releases.  It is fixed in the two releases since. ... View more

If you fit your model in PROC LOGISTIC, you can use the SCORE statement with the FITSTAT option to score a new data set using the fitted model and get the area under the ROC curve (AUC) along with several other statistics.  See the "Details: Scoring data sets" section of the the LOGISTIC documentation. Note that the Gini statistic equals 2*AUC-1. Or, you can use the predicted probabilities for the new data saved from the OUT= option of the SCORE statement in the PRED= option of the ROC statement in a subsequent run of PROC LOGISTIC to get Somers' D (which is the Gini statistic) along with the gamma and tau statistics (as well as the AUC again) as shown in this note.  ... View more

The note I referred to shows exactly the sort of comparisons you seem to want in your original post. Please look over the examples. The SLICE, LSMEANS, and LSMESTIMATE statements work the same way in any modeling procedure that supports them.   slice group*sex / sliceby=sex;  and slice group*sex / sliceby=group;   Note that you need to use PARAM=GLM (not REF) in the CLASS statement to use SLICE, LSMEANS, or LSMESTIMATE.  See the documentation of the SLICE and LSMEANS statements in your modeling procedure to see the additional options that are available. ... View more