Events and Meetings of Italian Statistical Society, Advances in Latent Variables - Methods, Models and Applications

Font Size: 
Specification of random effects in multilevel models: an overview with focus on school effectiveness
Carla Rampichini, Leonardo Grilli

Last modified: 2013-06-16


The analysis of highly structured data requires models with unobserved components (random effects) able to adequately account for the patterns of variances and correlations. An appropriate specification of the unobserved components is a key and challenging task. In this paper, we first review the literature about the consequences of misspecifying the distribution of the random effects for both estimation and prediction, then we outline the main alternatives and generalizations, also considering some issues arising in Bayesian inference. The relevance of suitably structuring the unobserved components is illustrated by means of an application exploiting a model with heteroscedastic random effects.

Full Text: PDF