【主讲】Stefan Hoderlein (Boston College)
【主题】The Triangular Model with Random Coefficients
【时间】2017年3月8日 (周三) 15：30-17：00
【地点】上海财经大学经济学院楼702室
【语言】英文
【摘要】The triangular model is a very popular way to allow for causal inference in the presenceof endogeneity. In this model, an outcome is determined by an endogenous regressor,which in turn is first caused by an instrument. In this paper, we study the triangular modelwith random coefficients and exogenous regressors in both equations. We establish nonidentificationof the joint distribution of the random coefficients, implying that counterfactualoutcomes are also not identified. This result continues to hold, if we confine ourselvesto the joint distribution of coefficients in the outcome equation or any marginal, except theone on the endogenous regressor. Identification continues to fail, even if we focus on meansof random coefficients (implying that IV is generally biased), or let the instrument enterthe first stage in a monotonic fashion. Based on this insight, we derive bounds on the jointdistribution of random parameters, and suggest an additional restriction that allows to pointidentify the distribution of random coefficients in the outcome equation. We extend thisframework to cover the case where the regressors and instruments have limited support, andanalyze semi- and nonparametric sample counterpart estimators in finite and large samples.Finally, we provide an application of the framework to consumer demand.
