ABSTRACT
Şiray et al. proposed a restricted Liu estimator to overcome multicollinearity in the logistic regression model. They also used a Monte Carlo simulation to study the properties of the restricted Liu estimator. However, they did not present the theoretical result about the mean squared error properties of the restricted estimator compared to MLE, restricted maximum likelihood estimator (RMLE) and Liu estimator. In this article, we compare the restricted Liu estimator with MLE, RMLE and Liu estimator in the mean squared error sense and we also present a method to choose a biasing parameter. Finally, a real data example and a Monte Carlo simulation are conducted to illustrate the benefits of the restricted Liu estimator.
Acknowledgments
The author is highly obliged to the editor and the reviewer for the comments and suggestions which improved the article in its present form. This work was supported by the National Natural Science Foundation of China (No. 11501072) and the Natural Science Foundation Project of CQ CSTC (grant no. cstc2015jcyjA00001).