Title: On the asymptotic properties and information criteria for misspecified generalized linear mixed models
Authors: Dalei Yu - Yunnan University of Finance and Economics (China) [presenting]
Xinyu Zhang - Academy of Mathematics and Systems Science, Chinese Academy of Sciences (China)
Kelvin Yau - City University of Hong Kong (Hong Kong)
Abstract: The problem of misspecification poses challenges in model selection. We study the asymptotic properties of estimators (including the estimators of fixed effects, variance component parameters and predictors of random effects) for generalized linear mixed models with misspecification under the framework of conditional Kullback-Leibler divergence. A conditional generalized information criterion is introduced, and a model selection procedure is proposed by minimizing the criterion. We prove that the proposed model selection procedure is asymptotically loss efficient when all the candidate models are misspecified. We also investigate the model selection consistency of the proposed model selection criterion. Numerical studies confirm the effectiveness of the proposed criterion.