Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 277 - SPEED: Biometrics and Environmental Statistics Part 1
Type: Contributed
Date/Time: Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
Sponsor: Biometrics Section
Abstract #322252
Title: Comparison of Test Statistics for Regression Coefficients in the OLS, Ridge, Liu, and Kibria-Lukman Linear Regression Model: A Simulation Study
Author(s): Sergio Perez-Melo* and Zoran Bursac and B.M. Golam Kibria
Companies: Florida International University and Florida International University and Florida International University
Keywords: Liu regression; Ridge regression; Kibria-Lukman regression; type I error rate; shrinkage parameter; empirical power

Ridge, Liu and Kibria-Lukman regression are methods that have been developed to minimize the multicollinearity problem for both linear and non-linear regression models. This paper proposes some new Ridge, Liu and Kibria-Lukman regression t-type tests of the individual coefficients of a linear regression model. A simulation study was conducted to evaluate and compare the performance of the tests with respect to their empirical size and power for different levels of correlation in the data. Our simulations allow us to identify which one of the proposed tests maintains type I error rates close to the 5% nominal level, while at the same time showing considerable gain in statistical power over the standard ordinary least squares (OLS) t-test. Our paper is the first of its kind in comparing the t-type tests for these different approaches to estimation. The results will be of value for applied statisticians and researchers.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program