On the Performance of Preliminary Ridge Regression Estimator Under the BLINEX Loss Function

• Activity Number: 612
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Risk Analysis
• Author(s): Judy Kleyn*+ and Andriette Bekker and Sollie Millard and Mohammad Arashi
• Companies: University of Pretoria and University of Pretoria and University of Pretoria and Shahrood University
• Keywords: Bounded linear exponential loss function ; Cobb-Douglas production function ; Preliminary test ridge regression estimator ; Ridge regression

Abstract

In this paper, we consider the multiple regression model in the presence of multicollinearity. We study the performance of the preliminary test estimator (PTRRE), when it is a priori suspected that some constraints may hold on the vector parameter space. The performance of the PTRRE is analysed by comparing the risk of some well-known estimators of the ridge parameter through an extensive Monte Carlo simulation study under the bounded linear exponential (BLINEX) loss. An application of the proposed theoretical results, when evaluating the Cobb-Douglas production function, is included and from the results it is clear that the bounded linear exponential loss function outperforms all other loss functions across all the proposed ridge parameters.