Estimation of the Transition Probabilities between HIV Disease Progression Categories
Open Access
Author:
Li, Shanglun
Area of Honors:
Interdisciplinary in Mathematics and Statistics
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
Le Bao, Thesis Supervisor Sergei Tabachnikov, Thesis Honors Advisor Murali Haran, Thesis Honors Advisor
Keywords:
HIV Markov Chain Maximum Likelihood
Abstract:
In this project, we investigate the effect of treatment, age, and gender to the progression of AIDS by using CD4 data. We fit a Markov chain model, and use the Maximum Likelihood approach to estimate the transition probability matrices. Some numerical analysis are applied on the resulting transition probabilities, such as comparisons of confidence intervals of transition probabilities across different subgroups.