Diagnostics for generalized Poisson regression models with errors in variables

Abstract

In this paper, we develop diagnostic methods for generalized Poisson regression (GPR) models with errors in variables based on the corrected likelihood. The one-step approximations of the estimates in the case-deletion model are given and case-deletion and local influence measures are presented. Meanwhile, based on a corrected score function, the testing statistics for the significance of dispersion parameters in GPR models with measurement errors are investigated. Finally, illustration of our methodology is given through numerical examples.

Keywords:

• generalized Poisson regression
• dispersion parameter
• local influence
• case-deletion model
• score test
• measurement error models
• corrected likelihood

Acknowledgements

We would like to thank an Associate Editor and referees for their helpful comments and suggestions that led to a significant improvement of the paper. This work is supported by NSFC (10671032) and the Grant for Young Teachers in Nanjing Agricultural University (KJ06036).