These "expected mean squares" can be used as the basis for estimation of the "variance components" σ and τ.

The type of statistical tests that are employed in multilevel models depend on whether one is examining fixed effects or variance components.

However, variance components can differ, as some groups are more homogeneous than others.

Therefore there three categories of family-based linkage analysis - strongly modeled (the traditional lod score model), weakly model based (variance components methods), or model free.

One method used to fit such mixed models is that of the EM algorithm where the variance components are treated as unobserved nuisance parameters in the joint likelihood.

Its application was originally to the estimation of variance components in random effects models.

In 1967 he and J.N.K. Rao published a maximum likelihood method for finding variance components in mixed models.

Random Effects: Random effects are the variance components that arise from measuring the relationship of the predictors to Y for each subject separately.

The second set of methods of estimation of heritability involves ANOVA and estimation of variance components.

They also allow exact inference on variance components of mixed models.