【Metrics Seminar】Lee Bounds with a Continuous Treatment in Sample Selection
2024/07/23
研討會日期 : 2024-07-23
時間 : 14:30
主講人 : Professor Ying-Ying Lee
地點 : Conference Room B110
主持人 : Professor Chu-Au Liu
演講者簡介 : Professor Ying-Ying Lee received her PhD from the University of Wisconsin-Madison in 2013. She is currently an Associate Professor at the University of California at Irvine. Her research fields of interest are econometric theory and empirical microeconomics.
演講摘要 : Sample selection problem occurs when treatment affects the researcher’s ability to observe an outcome, inaddition to the outcome itself. This paper generalizes Lee (2009)’s bounds for the average treatment effect ofa binary treatment to a continuous/multivalued treatment. We introduce a new concave-monotone assumptionon the treatment effect on selection. We evaluate the Job Crops program to study the causal effect of traininghours on wages. We further allow subjects with different pretreatment covariates to have either a positive ornegative treatment effect on selection. The estimation and inference theory utilize the orthogonal momentfunction and cross-fitting for double debiased machine learning.
