时间：2017年9月20日 上午9点开始

地点：邵逸夫科学馆401

题目：Testing Asymmetric Dependence: A Stochastic Process Approach

主讲人：宋晓军  北京大学光华管理学院助理教授

主讲人简介：宋晓军毕业于马德里卡洛斯三世大学经济系，获经济学博士学位,已经在《Journal of Business & Economic Statistics》， 《Econometric Reviews》等国际计量经济学杂志发表数篇文章，并有3篇学术论文在世界顶级计量杂志《Journal of Econometrics》的修改和等待发表中，曾获Dean’s Letter for Teaching Excellence教学奖，参加数十次国际研讨会议并发言，并曾志愿服务、义工奉献于多次国际会议。

Abstract: We propose new model-free tests for symmetric dependence between random variables. We consider the popular Cramer-von Mises and Kolmogorov-Smirnov-type test statistics based on the distance between positive and negative joint conditional exceedance distribution functions. These tests capture both linear and nonlinear dependence and do not require nonparametric smoothing. We derive their asymptotic distributions and establish the validity of a multiplier-type bootstrap that one can use in finite-sample settings. We also show that these nonparametric tests are consistent for any fixed alternative and they have non-trivial power for detecting local alternatives converging to the null at the parametric rate. A Monte Carlo simulation study reveals that the bootstrap-based tests control the size and have good power for a variety of data generating processes and different sample sizes. Finally, we provide an empirical application where we test the symmetric dependence between the S&P 500 daily return and the daily returns on 29 individual stocks. The results indicate that the symmetric dependence hypothesis is rejected for the majority of stocks under consideration.