Date of Award
5-13-2021
Degree Type
Thesis
Degree Name
Master of Science (MS)
Department
Mathematics and Statistics
First Advisor
Yichuan Zhao
Second Advisor
Jun Kong
Third Advisor
Jing Zhang
Abstract
In the thesis, we consider the Cox regression model. We develop the jackknife empirical likelihood (JEL), adjusted jackknife empirical likelihood (AJEL), mean jackknife empirical likelihood (MJEL), transformed jackknife empirical likelihood (TJEL) and (TAJEL) transformed adjusted jackknife empirical likelihood for the inference about the regression parameters. Additionally, the adjusted empirical likelihood (AEL), mean empirical likelihood (MEL), transformed empirical likelihood (TEL) and transformed adjusted empirical likelihood (TAEL) methods are developed. We compare methods under different distributions in terms of the coverage probability and average length of confidence interval for the regression parameter with simulation studies and three real data sets. The simulation analyses indicate that the MJEL, AJEL, and TAJEL methods are the best performing JEL methods while the MEL method was the best performing EL method. The real data analyses yielded results consistent with the simulation studies.
DOI
https://doi.org/10.57709/22762939
Recommended Citation
Drinkard, Lauren, "Jackknife Empirical Likelihood Methods for the Cox Regression Model." Thesis, Georgia State University, 2021.
doi: https://doi.org/10.57709/22762939
File Upload Confirmation
1