CONSISTENT VARIABLE SELECTION USING R2 CRITERION IN GENERALIZED LINEAR MODELS

In this article, we propose a consistent variable selection criterion based on coefficient of determination in generalized linear models. We establish its consistency property in the sense that asymptotically, model selected using our criterion coincides with the optimal model with probability one. A key advantage of our criterion is its applicability to the quasi-models. A simulation study was conducted to compare the performance of proposed criterion with the existing ones. A real data application is also presented.