|Saturday, February 22|
|PS3 Poster Session III & Continental Breakfast||
Sat, Feb 22, 7:30 AM - 9:00 AM
A Comprehensive Simulation Study On Some Ridge Regression Estimators and their Performances (302723)Shipra Banik, Independent University
*Golam BMG Kibria, Florida International University
Keywords: Linear Regression; MSE; Multicollinearity; Ridge Regression; Simulation Study
The estimation of ridge parameter is an important problem in the ridge regression, which is widely used to solve the multicollinearity problem. Many methods are available in literature for estimating the ridge parameter k. This paper considers a comprehensive study on 28 different available ridge estimators and proposed few estimators. A simulation study has been conducted and a mean square error criterion is used to compare the performances of the estimators. We observe that some of the selected ridge estimators performed well compared to the ordinary least square estimator and some existing popular ridge estimators. Finally, two numerical examples have been considered to illustrate the performance of the estimators.