题目：嘉宾：林丹瑜 教授（ University of North Carolina at Chapel Hill, USA）
时间：2014年6月27日下午3:30
地点：管理科研楼10楼1018会议室
单位：管理学院统计与金融系
摘要：
 Meta-analysis is widely used to synthesize the results of multiple studies. Although meta-analysis is traditionally carried out by combining the summary statistics of relevant studies, advances in technologies and communications have made it increasingly feasible to access the original data on individual participants.  We investigate the relative efficiency of analyzing original data versus combining summary statistics. We show that, for all commonly used parametric and semiparametric models, there is no asymptotic efficiency gain by analyzing original data if the parameter of main interest has a common value across studies; when the parameter of main interest follows a random-effect distribution, the maximum likelihood estimation of original data can be even less efficient than combining summary statistics. We demonstrate these theoretical results using both simulated and real data.