报告题目：Interval Estimation for Semiparametric Predictive Regression

报告人：蒋建成教授(University of North Carolina at Charlotte)

主办单位：统计学院

时间：2017年6月21日下午15:30-16:30

地点：大数据研究院201会议室

Abstract：Predictive regression is an important research topic in financial econometrics. To test the predictability, various methods have been proposed, but they suffer from a complicated asymptotic limit which depends on whether or not the predicting variable is stationary. In this talk we employ a nonlinear projection to deal with endogeneity of the state variable which results in a new semiparametric predictive regression model for describing the relationship between the state variable and the asset return. A weighted profile least square approach is used to estimate the parameters, and an empirical likelihood method is employed to test predictability. We establish the asymptotic distributions of the proposed estimators and show the Wilks theorem holds for the test statistic regardless of predicting variable being stationary or not. This provides a unifying method for constructing interval estimators of the coefficients of predicting variable. Simulations are conducted to demonstrate nice finite performance, and a real example is used to illustrate the use of the proposed method.

报告人简介：University of North Carolina at Charlotte数学与统计系副教授，主要从事生物统计、金融计量经济学、非参数统计、纵向数据等方面的研究。在The Journal of the American Statistical Association (JASA)、Annals of statistics、J. Roy. Statist. Soc. B国际著名统计期刊发表论文50余篇。