学术沙龙主题:Testing for the martingaledifference hypothesis in multivariate time series models
报告人:王国长 教授 暨南大学
邀请人：李本崇
报告时间:2022年4月28日(周四)；下午16:00—17:30
报告地点：腾讯会议ID：203613815
报告人简介：王国长，现任暨南大学经济学院统计学系教授、博士生导师。2012年毕业于东北师范大学best365亚洲版登录统计系，并取得统计学博士学位，2012-2014年在中国科学院应用所从事博士后研究工作，2017-2018年赴香港大学统计与精算系学术访问1年。主要研究方向为函数型数据分析、时间序列、充分性降维等，迄今为止在Journal of Econometrics, Journal of the Business & Economic Statistics, Statistica Sinica,Scandinavian Journal of Statistics等重要学术期刊接收和发表论文20余篇。主持国家社科基金一般项目和国家自然科学基金青年基金项目各1项，主持博士后面上项目和广东省自然科学基金面上项目各1项。任中国现场统计研究会资源与环境统计分会常务理事；中国旅游大数据协会，理事，副秘书长；广东省现场统计协会常务理事，副秘书长。
报告摘要:This paper proposes a general class of tests to examinewhether the error term is a martingale difference sequencein a multivariate time series model with parametric conditional mean. These new tests are formed based on recently developedmartingale difference divergence matrix (MDDM),and they provide formal tools to test the multivariate martingale hypothesis in the literature for the first time.Under suitable conditions, the asymptotic null distributions of these MDDM-based tests are established. Moreover, theseMDDM-based tests are consistent to detect a broad class of fixed alternatives, and have nontrivial power against local alternatives of order $n^{-1/2}$, where $n$ is the sample size.Since the asymptotic null distributions depend on the data generating process and the parameter estimation,a wild bootstrap procedure is further proposed to approximate the critical values of these MDDM-based tests, and its theoreticalvalidity is justified. Finally, the usefulness of these MDDM-based tests is illustrated by simulation studies and one real data example.
