Intra-cluster correlation as serial vs. equicorrelation in a hierarchical linear model

Abstract

This paper proposes a simple hierarchical model and a testing strategy to identify intra-cluster correlations. Intra-group correlations are modeled as a combination of nested random effects and serially correlated error components in a hierarchical model. A Neyman C(α) framework is used to derive LM-type tests to identify the appropriate level of clustering and the type of intra-group correlation.