Structural Instrumental Variable Estimator for Dynamic Treatment Effects: Spanking Effect on Behavio
2009/05/05
研討會日期 : 2009-05-05
時間 : 14:30
主講人 : Professor Myoung-Jae Lee
地點 : C103
演講者簡介 : Professor Myoung-Jae Lee為Ph.D. in Economics，University of Wisconsin-Madison (1989)。目前為Professor，Korea University。其主要研究領域為Econometrics and Statistics、Labor and Health Economics。
演講摘要 : Finding the effects of multiple sequential treatments on a response variable measured at the end of a trial is difficult, if some treatments are affected by interim responses; e.g., assessing the effects of sequential spankings on child behavior when parents adjust their interim spanking levels depending on interim behaviors. A headway, ‘G estimation’, has been made in 1980’s generalizing the usual static single-treatment effect analysis under ‘no unobservable confounders’ or ‘selection on observables’. But G estimation is not easy to implement. In this paper, firstly, we propose a simpler alternative to G estimation–instrumental variable estimator (IVE) for a linear structural model–and show that our proposal and G estimation identify the same effect under some assumptions. Secondly, we explore the relation between our proposal and Granger causality to show that our approach is more general, although the two become equivalent for testing non-causality under a stationarity-type assumption. Thirdly, our approach and G estimation are applied to find the effects of spanking on child behavior. We find that mild spanking at early years reduces child behavior problems later, which seems to differ from most findings in the psychology/education literature.
