A Comparative Study of Robust Estimators for Poisson Regression Model with Outliers

Authors

Mohamed Reda Abonazel, Department of Applied Statistics and Econometrics, Faculty of Graduate Studies for Statistical Research, Cairo University, Giza, EgyptFollow
Omnia Mohamed Saber

Author Country (or Countries)

Egypt

Abstract

The present paper considers Poisson regression model in case of the dataset that contains outliers. The Monte Carlo simulation study was conducted to compare the robust (Mallows quasi-likelihood, weighted maximum likelihood) estimators with the nonrobust (maximum likelihood) estimator of this model with outliers. The simulation results showed that the robust estimators give better performance than maximum likelihood estimator, and the weighted maximum likelihood estimator is more efficient than Mallows quasi-likelihood estimator.

Digital Object Identifier (DOI)

http://dx.doi.org/10.18576/jsap/090208

Recommended Citation

Reda Abonazel, Mohamed and Mohamed Saber, Omnia (2020) "A Comparative Study of Robust Estimators for Poisson Regression Model with Outliers," Journal of Statistics Applications & Probability: Vol. 9: Iss. 2, Article 8.
DOI: http://dx.doi.org/10.18576/jsap/090208
Available at: https://digitalcommons.aaru.edu.jo/jsap/vol9/iss2/8