A New Ridge-Type Estimator for the Gamma regression model

Abstract

When there is collinearity among the regressors in gamma regression models, we present a new two-parameter ridge estimator in this study. We look into the new estimator's mean squared error characteristics. Additionally, we offer several theorems to contrast the new estimators with the current ones. To compare the estimators under various collinearity designs in terms of mean squared error, we run a Monte Carlo simulation analysis. We also offer a real data application to demonstrate the usefulness of the new estimator. The results from simulations and actual data reveal that the proposed estimator is superior to competing estimators.

Keywords

• Multicollinearity
• ridge estimator
• gamma regression model
• Liu-type estimator
• Monte Carlo simulation