Estimation of structural parameters in crossed classification credibility model using linear mixed models

In this paper, the linear mixed model is used under the Dannenburg's two-way crossed classification model. Maximum likelihood (ML) and restricted maximum likelihood (REML) methods are employed to estimate the structural parameters with both independent and exchangeable error structures. Evidenced by results of simulation studies, the proposed linear mixed effects estimators appear to outperform those given by Dannenburg with both independent and exchangeable error structures.