Weighted high leverage collinear robust ridge estimator in logistic regression model
The combination of high leverage points and multicollinearity problem occurs frequently in logistic regression model. Methods that successfully address these problems separately are not effective for the combined problems. A robust logistic ridge regression (RLR) which incorporates the weighted Bian...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Pakistan Journal of Statistics
2018
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/74432/1/2018PJS.pdf http://psasir.upm.edu.my/id/eprint/74432/ https://www.pakjs.com/wp-content/uploads/2019/09/34105.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universiti Putra Malaysia |
| Language: | English |
| Summary: | The combination of high leverage points and multicollinearity problem occurs frequently in logistic regression model. Methods that successfully address these problems separately are not effective for the combined problems. A robust logistic ridge regression (RLR) which incorporates the weighted Bianco and Yohai (WBY) robust estimator with fully iterated logistic ridge regression (LR) is proposed to rectify the combined problems of high leverage points and multicollinearity in a data. A numerical example and simulation study are presented to compare the performance of the RLR with the ML, the WBY, and the LR estimators. Results of the study indicate that the RLR outperforms the established estimators for the combined problems. |
|---|