学院动态

学术观点

经济学院2021年数量经济学术讲座

文章来源: 发表时间:2021-04-07 15:01:45点击次数:

主讲人:涂云东

单位:北京大学光华管理学院

题目:Averaging Factor Models with Nonparametrically Targeted Predictors

时间:2021年4月9日(周五)9:00--10:30am

地点:经济学院407

讲座内容:This paper proposes a new factor model that can summarize the nonlinear information content contained in observed predictors into factor formation and subsequent factor-augmented predictions. Nonparametric sieve method is adopted to approximate the nonlinear functional space of the predictors, with the sure independence screening (SIS) procedure used to target the relevant nonlinear predictors to reduce predictor dimension and improve prediction efficiency. Theoretical results including the sure screening property of SIS and the asymptotic normality of the predictive estimator are established. Further refinements are achieved through the extended Bayesian Information Criterion (EBIC) and jackknife model averaging (JMA) in the prediction accuracy of the approximating factor model. The screening consistency of EBIC and the asymptotic optimality of JMA are then proved. Finally, numerical results demonstrate the nice performance of our proposed methods.

主讲人简介:涂云东,北京大学光华管理学院商务统计与经济计量系和北京大学统计科学中心联席副教授,研究员。入选首批“日出东方”光华青年人才,北京大学优秀博士学位论文指导教师,教育部“长江学者奖励计划”青年长江学者。2012年获美国加州大学河滨分校经济学博士学位,同年6月加入北大光华。30余篇学术论文发表在Journal of Econometrics, Econometric Reviews, Journal of Business and Economic Statistics, Oxford Bulletin of Economics and Statisitics,Statistica Sinica,Journal of Empirical Finance,Computational Statistics and Data Analysis, 《系统工程理论与实践》等国际国内知名专业杂志,并为多个专业学术杂志和自然科学基金匿名评审。

微信公众号