Your RANDOM statemen specification is unusual. That might be causing the nonconvergence issue. Can you send in your sample data? And explain what RESP_ID and PANT_ID are? Thanks, Jill ... View more

By "outcome values", do you mean predicted values? if so, you would add the OUTP=preddata option in the MODEL statement in PROC MIXED. Then the preddata data set would contain the predicted values. Hope this helps, Jill ... View more

Non-est lsmeans are often caused by data with missing cells. If you do a proc freq - proc freq; tables innocyear*harvestyear*TreeSp; run; do you see 0 count for one or more cells? If so, that explains it. You either combine some of the categories so no missing cells are present, or change your model, for example, taking out this three-way interaction term. You do not have two-way interactions in the model? -- something to think about adding.... Thanks, Jill ... View more

One approach might be to just use one observation for each Pen.   ... View more

It seems to me that for every Pen, you have two observations. The values for these two observations for each of Pref1-Pref5 always add up to 100. That might be a problem. You might want to reconsider how you want to arrange your data and set up your model. Thanks, Jill ... View more

Unfortunately there are no good ways that I am aware of to do what you asked for with your model. Thanks, Jill ... View more

For the long run time issue, you might try the following approach to see if it helps: proc sort data=tre; by id; run; proc glimmix data=tre; class l2 period; model count=period knt1-knt3 / dist=poisson link=log offset=offset; random intercept / subject=l2; random intercept / subject=id; run; For the convergence issue, you might want to double check your offset variable -- it should be the log of the measure of exposure.  Jill ... View more

Crossed random effects models are known to be resource intensive. I am not surprised that it took a very long time. Can you send in your program? Also how many levels in Ward and how many levels in MRN? Thanks, Jill ... View more

crossed random effects are specified like something like the following-- random int / subject=ward; random int / subject=mrn; (if data allows, sometimes you might also add random int / subject=ward*mrn; but my experience indicates that this random effect is often not supported adequately by data). Hope this helps, Jill ... View more

This statement -- random intercept / subject=oelat(subjid) type=cs; models a random intercept. So for each subject, there is only one random effect -- intercept.  TYPE=CS is too much. Use TYPE=VC for the random intercept. Hope this helps, Jill ... View more

Try using PARMS statement to enter you own starting values. This might take several trial and error. If that does not help, can you send in your data? Thanks, Jill ... View more

What happens if you add METHOD=LAPLACE option in the PROC GLIMMIX statement? Also, you might want to spell out: dist=binary in the MODEL statement. Thanks, Jill ... View more

Short answer: PROC CALIS does not have a keyword or an option for explained variance computations. Long answer: PROC  CALIS uses an approach that is philosophically different from exploratory factor analysis (PROC FACTOR), and so variance explained by factors is not usually considered a very important concept. In PROC FACTOR, factors are extracted and so you almost always want the variance explained information. In PROC CALIS (or confirmatory factor model), factors are hypothesized and the variance explained by the factors is not critical for selecting the number of factors. You can compute that information from CALIS: (1) communalities of variables in FACTOR = R-squares in the Squared Multiple Correlations of CALIS. (2) For the variance explained by the factors, you can do this: (a) Use ODS output to capture the RotatedLoadings table into a SAS data set; (b) The (unweighted) variance explained by the factors is the sum of squares of the loadings in columns. Note that these steps apply to the orthogonal solutions only. For oblique factors, these steps are not applicable (however, you cannot separate the explained variance by oblique factors anyway).   Hope this helps, Jill ... View more

Convergence issues in PROC NLMIXED are not straightforward to deal with, unfortunately. You might want to start with a simpler model, for example, take out random effects, and see how it goes. Sometimes you might want to play with different starting values in the PARMS statement, and this takes trial and error, and luck... Good luck, Jill ... View more