喻园管理论坛2024年第20期(总第952期)

演讲主题: Doubly Robust Estimation and Inference in a High Dimensional Transformation Model

主 讲 人: 张   健，南开大学经济学院数量经济研究所助理教授

主 持 人: 蔡   俊，会计与财税系讲师

活动时间: 2024年3月22日（周五）10:00-11:30

活动地点: 管院大楼119室

主讲人简介：

张健，南开大学经济学院数量经济研究所的助理教授。本科毕业于南开大学经济学专业，2022年于美国University of Wisconsin-Madison取得经济学博士学位，师从Bruce Hansen教授，目前主要从事高维非参数计量和半参数统计的研究。已有一篇论文发表在Econometric Theory，多篇论文在投。同时，张健也是国际期刊Review of Economics and Statistics的匿名审稿人。

活动简介：

This paper studies estimation and inference transformation model in the presence of a high dimensional set of control variables. In the study, I consider a generalized form of transformed model, which includes traditional transformed model, binary choice model and generalized accelerated failure time model as special cases. I include both low dimensional covariates of interest and high dimensional control variables in this model. The estimation of high dimension nuisance parameters could lead to substantial bias, and thus incorrect inference on parameters of interest. I provide a double-machine learning estimator to reduce this substantial bias, and obtain a square root n consistent and asymptotically normal results. According to simulation study, I compare the performance of our estimator with the classical estimator based on average partial derivatives, it turns out that our estimator has less bias and provides correct inference results. Based on this coefficient estimator, I also propose an two-step estimation of average partial effects, which utilizes double de-biased method and the model structure. Finally I also use 2018 ACS data to demonstrate the performance of my estimator.