Masters candidate in Biostatistics, Shengwei Zhang, will present:

“A Bayesian Hierarchical Model for Meta-Analysis of Demand Curves”

Plan B Advisers: Haitao Chu and Xianghua Luo

Abstract: Meta-analysis is a frequently used method to combine and contrast data from multiple independent studies. To appropriately consider the heterogeneity between studies, Bayesian hierarchical model is increasingly utilized in meta-analysis. Behavioral science is a popular area of the application of meta-analysis. In the realm of behavioral science, an exponential demand curve generated from cigarette purchase task (CPT) data describes how demand of cigarette varies as a function of price. In this paper, we propose a Bayesian hierarchical model for meta-analysis of demand curves to properly account for between-study heterogeneity. With individual participant data, the model is able to estimate the study- and population-level parameters simultaneously without normal assumptions as conventional random-effects meta-analysis method. We apply the proposed model to a meta-analysis with baseline individual CPT data from 6 studies and compare the results from the proposed model and conventional random-effects meta-analysis method. We conducted simulation analysis to demonstrate the performance of the proposed approach and discuss the benefits of using the Bayesian hierarchical model for meta-analysis of demand curves.

Refreshments will be served prior to the presentation.