When you add a repated statement, you are perform GEE, which involves a quasi-likeliood (not a true likelihood or true distribution). Thus, the deviance (which is based on a true likelihood) is not printed. ... View more

Yes, most sources say that MIXED was introduced in 1992.It was greatly changed, expanded, and improved over the following few years. ... View more

I still don't understand your question. Sorry. The SE is the measure of variability (uncertainty) for the estimated parameter of a covariable. It is printed right next to the parameter estimate in the solution table (just use the  "/ s" option on the model statement. You can use to get confidence intervals. One does not multiply the SE by anything related to df. ... View more

You didn't show code, so it is not clear what you are asking. If you want SEs for means, put in LSMEANS statments, or use ESTIMATE or LSMESTIMATE statements to get SEs for contrasts of means. This is all automatic. If you want a SE for each predicted observation, then you use an output statement (in GLIMMIX), or an outp= or outm= option in the model statement of MIXED.   With a mixed model, there are multiple sources of variability (more than one variance). Thus, there is no such thing as a CV.   ... View more

This makes a difference too, but this is a different issue from the original question. The so-called empirical estimate of the var-cov matrix of beta-hat is also known as the sandwhich estimate, and it removes (or tries to remove) the effect of certain aspects of the model choice on the estimate. It is sometimes called a robust estimate.   One gets the empical SEs by default with GENMOD (in GLIMMIX, it is not the default). To get the model-based SEs (actually, the SEs based on the model working correlation matrix), one puts MODELSE on the REPEATED statement.   ... View more

The regular SEs are based on the fitted model, including the working correlation matrix that is used in the GEE approach (i.e., the quasi-likelihood approach). They are consistent if the model and working correlation matrix is correctly specified. The empirical SEs are consistent even if the working correlation matrix is misspecified. Sometimes the empirical SEs are called robust SEs. The formulas are given in the Parameter Estimate Covariances sub-section of the Generalized Estimating Equations section. http://support.sas.com/documentation/cdl/en/statug/67523/HTML/default/viewer.htm#statug_genmod_details29.htm   ... View more

With the standard GLM parameterization, the intercept is the last level of the factor (level 6 in your case). You can take out the intercept (noint) and the six levels have estimates, with a direct interpretation. That is fine. However, the type III test for the factor is not testing the equality of the parameters (it is now testing the equality of the parateters all to 0). That is a big difference.   Using the default parameterization, you can use estimate statements to get the estimates for each level of the factor. Generic example: estimate '1' int 1 f 1 0 0 0 0 0; estimate '2' int 1 f 0 1 0 0 0 0; and so on. Here "f" is my generic name for your factor. ... View more

To be clear, use: proc mixed data=simulateddata(where=(size=5)); ...     ... View more

Be very careful with Laplace or Quadrature with binary or binomial data. If any of the random effects are 0, then everything blows up (without any warnings). Many of the standard errors of LSMEANS are 0 or go to zero, which means you get meaningless results. The Type III test could be extremely inlfated. You must remove any random effects with a 0 variance to get sensible results. THis is not an issue with RSPL estimation.    Early work suggested that RSPL could give biased results with binary response, so many have suggested Laplace or quadrature. However, recent work by Stroup shows that this is not always the case. My view, based on the most recent evidence, that there is no clear-cut best estimation method. But I do lean towards Laplace for purely binary repsonses. Just make sure you don't have any zero variances. ... View more