الفهرس 
Biased estimators for logistic panel data models
Glossary of NotationOnly 14 pages are availabe for public view
 |  | from | 181 |  |  |
| | | from | 181 | | |
Abstract
In this thesis, a biased estimator is proposed to combat multi-collinearity in the
logistic panel data regression model. This study aims at comparing the performance of
biased estimators with Maximum Likelihood was performed using the mean square error
criterion by applying to Monte Carlo simulation data and medical data analysis
(application study). The proposed estimator is a general estimator which includes other
biased estimators, such as the ridge estimator and the Liu estimator as special cases.
Furthermore, a Monte Carlo simulation study is given to illustrate some of the
theoretical results. Simulation results demonstrate that ridge logistic panel parameter is
more efficient than methods.
An application is also presented to assess the performance of the proposed ridge
estimators. The most significant factors that affect delayed completion of adjuvant
chemotherapy in patients with breast cancer. The study results show that the biased
estimator is more efficient and better than ML estimators. Moreover, we find that there are
very influential factors that affected delayed completion of adjuvant chemotherapy such as
Body Surface Area (BSA), Hemoglobin (HGB), Alanine Transaminase (ALT) and
Creatinine (SRCR). Finally, both simulation and application results, the proposed
estimator is much better than the ML estimator with respect to the MSE criteria.