报告时间：6月29日上午9:00——11:00
报告地点：统计楼105报告厅
报告一：
报告人：朱烨莹， University of Waterloo 助理教授，华师大校友
报告题目：Penalized regression procedures for variable selection in the potential outcomes framework
报告摘要：
A recent topic of much interest in causal inference is model selection. In this article, wedescribe a framework in which to consider penalized regression approaches to variable selectionfor causal effects. The framework leads to a simple “impute, then select” class of proceduresthat is agnostic to the type of imputation algorithm as well as penalized regression used. Italso clarifies how model selection involves a multivariate regression model for causal inferenceproblems, and that these methods can be applied for identifying subgroups in which treatmenteffects are homogeneous. Analogies and links with the literature on machine learning methods,missing data and imputation are drawn. A difference LASSO algorithm is denied, along withits multiple imputation analogues. The procedures are illustrated using a well-known right heartcatheterization dataset.
报告二：
报告人：Naitee Ting，Fellow of ASA, currently a Sr. Principal Biostatistician in the Biometrics Department of Boehringer-Ingelheim Pharmaceuticals Inc. (BI).
报告题目：What is a Clinical Trial？
报告摘要：
Life expectancy in the US back around 1900 was in the upper forties. It increased up to around upper 70’s in year 2000. Modern science was able to bring life expectancy up by about 30 years within one century. There are many factors to help improve life expectancy – reduction of infant mortality, improvement in the hygiene system, invention of penicillin, … and many other important factors. Each of these factors contributes a few years of this 30 year increase in life expectancy. However, there should be at least 10 to 15 years of such an increase come from randomized, controlled, double-blind clinical trials. This presentation introduces the concept of randomization, the considerations in selection of controls, and blinding.