Title（标题）：Statistical Models for Estimating HIV Prevalence and the At-risk Population Size

Speaker（汇报人）：Mr. Le Bao（Ph.D candidate）

Department of Statistics

University of Washington, USA

Time（功夫）： 2011年4月14日（周四）下午2:00-3:00

Place（地址）： 北京大学性命科学学院610教室

Abstract（提要）：Estimation and Projection Package (EPP) is used to estimate and project adult HIV prevalence and incidence from surveillance data. I will present an Incremental Mixture Importance Sampling (IMIS) algorithm which improves the sampling efficiency of EPP while retaining its essential simplicity and transparency. It gives a good picture of the uncertainty when the posterior distribution tends to be multimodal and to feature nonlinear sharp ridges. For low-level and concentrated HIV epidemics, where transmission is not sustained out-side at-risk populations, it is more important to estimate the size of populations that are most likely to acquire and transmit HIV. I will present a Bayesian hierarchical model for developing local and national size estimates of HIV at-risk populations and assessing the uncertainty of the estimates. It utilizes multiple data sources including mapping data, surveys, interventions, capture-recapture data, estimates and expert opinions. The model is used to estimate the numbers of injecting drug users in Bangladesh and yields satisfactory results.