A New Modified Liu Ridge-Type Estimator for the Gamma Regression Model

Abstract

The Gamma Regression Model (GRM) is a special form of the generalized linear model (GLM), where the response variable is positively skewed and well fitted to the gamma distribution. The most popular technique for estimating GRM coefficients is maximum likelihood (ML) estimation. The ML estimation method performs better if there is no correlation between the explanatory variables. It is known that the variance of the maximum likelihood estimator of the gamma regression coefficients is impacted in situations when the explanatory variables are correlated. Based on Aslam and Ahmad’s Modified Liu-Ridge-Type (MLRT) estimator, which has been demonstrated to lessen the effects of multicollinearity in the linear regression model, the paper presented a New Modified Liu Ridge-Type Estimator for the Gamma Regression Model (GMLRT) to address the problem of multicollinearity. Simulation and real-world application results show the superiority of the GMLRT estimator using the mean square error criterion.