Significance Testing

As with ordinary least squares regression or logistic regression, we can consider significance tests for individual estimates, such as intercepts, slopes, and their variances, as well as whether the full model accounts for a significant amount of variance in the dependent variable. In between, there is also the possibility of determining whether of subset of predictors contribute significantly. Aside from these fixed effects, we also can test the variance components or random effects (variance of intercepts, variance of slopes, or covariances among them) for significance. Unfortunately, there are several considerations for testing either fixed or random effects that make this an all too complicated topic.