(5-5PM2:30) Large System of Seemingly Unrelated Regressions:

报告名称：Large System of Seemingly Unrelated Regressions: A PenalizedQuasi-Maximum Likelihood Estimation Perspective

报 告 人：范青亮（厦门大学 助理教授）

时      间：2017年5月5日（周五）下午14:30-16:00

地      点：岭南堂汪道涵会议室

语      言：中文+英文

Abstract:

In this paper, we propose using a shrinkage estimator, penalized quasi-maximum likelihoodestimator (PQMLE), to estimate a large system of equations in seemingly unrelated regressions model where the number of equations is large relative to sample size. We develop the asymptoticproperties of the penalized quasi-maximum likelihood estimator for both error covariance matrixand model coefficients. In particular, we derive the asymptotic distribution of the coefficient estimatorand the convergence rate of the estimated covariance matrix in terms of Frobenius norm.The model selection consistency property of the covariance matrix estimator is also established.Simulation results show that when the number of equations is large relative to sample size and theerror covariance matrix is sparse, the penalized likelihood estimator performs much better thanconventional estimators. As an illustration, we apply the PQMLE to the study of state level publiccapital returns in the United States.

报告人介绍：

        范青亮，厦门大学王亚南经济研究院和经济学院助理教授，博士生导师。2012年毕业于美国北卡州立大学，获得经济学博士学位。主要致力于计量经济学理论与实证方面的研究，大数据分析等。目前已在国际国内顶尖学术期刊如Journal of Econometrics, Journal of Business and Economic Statistics等发表多篇论文。研究得到国家自然科学基金面上项目、青年项目，福建省自然科学基金面上项目等资助。

报告人主页：michaelqfan.weebly.com

个人简介：/UploadFiles/xsbg/2017/4/201704271653040612.pdf