Modified cross-validation procedure in selection shrinkage parameter of Poisson ridge regression model

Abstract

Poisson regression model is the standard statistical method for analyzing count data. Its parameters are usually estimated using the maximum likelihood (ML) method. However, the ML method is very sensitive to multicollinearity. Ridge estimator was proposed in Poisson regression model. The choice of the ridge shrinkage parameter is critical. Cross-validation method is a widely adopted method for shrinkage parameter selection. However, cross-validation method suffers from instability in determining the best shrinkage parameter. To address this problem, a modification of the cross-validation method is proposed. Simulation and real data example results demonstrate that the proposed method is outperformed by cross-validation and generalized cross-validation methods.