报告人：宋晓军

        北京大学光华管理学院

主持人：彭一杰 特聘研究员

时  间：2019年6月3日 下午3:00

地  点：工学院一号楼212会议室

报告内容摘要：

This article proposes new nonparametric tests to check the adequacy of the conditional mean specification for multiplicative error models (MEM) satisfying Markov structures. Our tests are based on a novel projection device, which kills "parameter estimation uncertainty" and also motivates a convenient multiplier bootstrap procedure to simulate critical values of the proposed tests as accurately as desirable. In addition, our tests are fully data-driven, do not require kernel function or user-chosen tuning parameters such as bandwidths, and are able to detect a broad class of local alternatives converging to the null at the parametric rate $n^{-1/2}$, with $n$ the sample size.

报告人简介：

Xiaojun Song currently is an Assistant Professor in Department of Business Statistics and Econometrics in Guanghua School of Management at Peking University. He got his PhD in Economics from Universidad Carlos III de Madrid, Spain. His main research interests are nonparametric/semiparametric methods, hypothesis testing, bootstrap, etc. He has published in journals like Journal of Econometrics, Journal of Business and Economic Statistics, etc.