Events and Meetings of Italian Statistical Society, Statistics and Demography: the Legacy of Corrado Gini

Font Size: 
New Methods to Estimate the Biasing Parameters Used in Liu-type Logistic Regression: An Application to Data of Bankruptcy of Turkish Banks during the Asian Financial Crisis
Yasin Asar, Adnan Karaibrahimo─člu

Last modified: 2015-09-05

Abstract


It is well known that the variance of the maximum likelihood estimator is affected negatively in the presence of high correlation among the explanatory variables called the multicollinearity problem. In this study, we propose some new estimators to estimate the parameters used in the Liu-type logistic estimators in order to decrease the variance and make stable inference. We conduct a Monte Carlo simulation to evaluate the performance of the proposed methods. An application of real data example which is related to the bankruptcy of Turkish Banks during the Asian financial crisis is illustrated to show the benefits of new methods. According to the results, new methods work well and recommended to the researchers.

Full Text: Untitled