A B-spline Hierarchical Model on HIV Prevalence Data
Open Access
Author:
Tang, Yuan
Area of Honors:
Mathematics
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
Le Bao, Thesis Supervisor Luen Chau Li, Thesis Honors Advisor
Keywords:
b-spline hierarchical model HIV prevalence mixed model Bayesian inference Markov Chain Monte Carlo sparse and imbalanced data
Abstract:
Due to the paucity of reliable information on the incidence of HIV in most countries, sentinel surveillance systems for HIV were designed to provide information on prevalence trends, e.g. the percentage of HIV positive cases is often estimated among antenatal clinic (ANC) patients. In recent years, surveillance sites have been established which allow countries to look into the epidemics at a finer scale, e.g. at sub-national level or sub-population level. An important technical barrier is that the availability and quality of the data vary widely from area to area, and many areas lack data for deriving stable and reliable results. In countries with low-level and concentrated epidemics, HIV has spread rapidly in several high risk groups, but is not well established in the general population. Fewer data are available for those sub-populations due to the stigmatized nature of these sub-populations in many countries. To improve the accuracy of the results in areas or high risk groups with little data, we propose a Bayesian hierarchical model that utilizes information more efficiently by assuming similarity of HIV prevalence across areas and sub-populations.