Improving the Estimates of Variance Ratios and BLUPs of Mixed-Effects Models

• Activity Number: 239
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Samaradasa Weerahandi*+ and Malwane Ananda
• Companies: Pfizer and University of Nevada, Las Vegas
• Keywords: Random Effects ; Best Linear Unbiased Predictor ; Generalized Estimate ; ML ; REML

Abstract

Lately Mixed Models are heavily employed in analyses of promotional tactics as well as in design and analysis of data from clinical trials. The Best Linear Unbiased Predictor (BLUP) in Mixed Models is a function of the variance components and they are typically estimated using conventional MLE based methods. It is well known that frequently the estimate of the factor variance becomes zero or negative. In such situations, ML and REML either do not provide any BLUPs or random effects all become practically zero. Moreover, such estimates are not admissible.

In this article we proposed a class of estimators that do not suffer from the negative variance problem, while improving upon existing estimators. The MSE superiority of the prosed estimator is illustrated by a simulation study.