Are you asking how to compare two nested models that you've estimated with the MIXED procedure? If so, you'll have to "roll your own" likelihood ratio test by getting the difference in the -2LL values for the two models. According to the null hypothesis (i.e., no improvement in the fit of the fuller of the two models), that difference is distributed as chi-square with df = the difference in the number of parameters for the two models.
If that is not what you're asking, please clarify the question.