A comparative study of alternative estimators for the unbalanced two‐way error component regression model

Abstract

This paper considers the unbalanced two-way error component modelstudied by Wansbeek and Kapteyn (1989). Alternative analysis of variance(ANOVA), minimum norm quadratic unbiased and restricted maximum likelihood(REML) estimation procedures are proposed. The mean squared errorperformance of these estimators are compared using Monte Carlo experiments.Results show that for the estimates of the variance components, thecomputationally more demanding maximum likelihood (ML) and minimumvariance quadratic unbiased (MIVQUE) estimators are recommended,especially if the unbalanced pattern is severe. However, focusingon the regression coefficient estimates, the simple ANOVA methodsperform just as well as the computationally demanding ML and MIVQUEmethods and are recommended. Copyright Royal Economic Society, 2002