Performance of Kibria, Khalaf, and Shurkur's methods when the eigenvalues are skewed

ABSTRACT

Many methods are available for the estimation of ridge regression parameter in literature. This article considered some of these estimators and also proposed some new methods that take care of the skewed eigenvalues of the matrix of explanatory variables. Simulated results obtained indicate that when the sample size increases, the Prediction Sum of Squares (PRESS) value decreases as the value of the correlation coefficient becomes large. One of the proposed methods outperforms all the other existing and proposed methods considered in terms of PRESS values. A numerical example with six explanatory variables was used to compare the performance of these estimators.

KEYWORDS:

• Eigenvalues
• Ridge parameter
• Shrinkage
• Skewness

MATHEMATICS SUBJECT CLASSIFICATION:

• 62J07

View correction statement:

Corrigendum