The semiparametric and nonparametric literature has seen recent developments in estimating econometric models when regressors are endogenous. In the first part of the lecture, we will introduce the endogeneity problem in a variety of econometric models, covering traditional linear models and general nonparametric and nonseparable models. In particular, we will give an overview of nonparametric identification and estimation methods that have been developed in the literature, with the focus on two prominent approaches: instrumental variables and control functions. The second part of the lecture will deal with a binary response model with endogenous regressors, which can be viewed as an important example of nonseparable models using control functions to address endogeneity problems. For this specific model, identification, estimation procedures, and asymptotic properties of the proposed estimator will be introduced.
講者介紹:
廖仁哲教授為美國 University of Wisconsin-Madison 經濟學博士,目前任職於中央研究院經濟研究所。研究領域為經濟計量,詳細期刊論文著作請參閱廖教授個人網頁。