Abstract

In this paper, we discuss how to model the mean and covariance structures in linear mixed models (LMMs) simultaneously. We propose a data driven method to model covariance structures of the random effects and random errors in the LMMs. Parameter estimation in the mean and covariances is considered by using EM algorithm, and standard errors of the parameter estimates are calculated through Louis’(1982) information principle. Kenward’s (1987) cattle data sets are analyzed for illustration, and comparison to the literature work is made through simulation studies. Our numerical analysis confirms the superiority of the proposed method to existing approaches in terms of Akaike information criterion.

(This is a joint work with Dr. Jianxin Pan, Professor of Statistics, The University of Manchester)

About the Speaker

FEI Yu, National Second-Class Professor, PhD., Postdoctoral Fellow of Manchester University, Distinguish Professor, Doctoral Supervisor and Executive Vice President of the College of Statistics and Mathematics of Yunnan University of Finance and Economics. He is known as Yunling scholar of Yunnan province, Yunnan Provincial Middle-Young Academic and Technical Leader, Yunling Distinguished Teacher and Yunnan Provincial Distinguished Teacher. He also works as an Executive Director of Commerce Statistical Society of China and the National Society of Economic Mathematics and Management Mathematics, Vice President of Yunnan Provincial Applied Statistics Association, and communication review expert of National Natural Science Foundation Project and National Social Science Foundation Project. Professor Fei’s research interests include statistical theory and methods, applied statistics, data mining and econometric analysis. He has published more than 30 papers in Technometrics、Communications in Statistics—Theory and Methods、Statistical Papers、Energy、International Journal of Heat and Mass Transfer and Statistical Research. He has also published two monographs and four textbooks. Professor Fei has held three National Natural Science Fund Projects and received ten provincial and ministerial awards.

Poster