报告题目：Robust Estimation for the Number of Common Factors in the Approximate Factor Models

报告人：吴鑑洪教授（浙江工商大学统计与数学学院）

报告时间：2015年5月12日（周二）下午2:00-3:00

报告地点：数学楼二楼学术报告厅

摘要：This paper proposes a robust estimation method for the number of common factors in the approximate factor models. The new estimation is based on the ratio values of some transformation function of the eigenvalues and the use of the so-called ridge-type parameter. Under some mild conditions, the estimation can be proved to be consistent.It can be shown that the new method enjoys some desired properties as follows: it is available for the case in which some factors are dominant or even non-stationary; it needs not the assumption of finite number of factors, that is, the true number of factors can be allowed to increase with the sample size; it needs not the pre-specification of the possible maximum number $kmax$. Monte Carlo simulation is carried out for illustration.

报告人简介：吴鑑洪教授，现任浙江工商大学统计与数学学院统计科学研究所副所长，博士生导师，中国人民大学应用统计研究中心兼职研究员。现为浙江省中青年学科带头人（统计学），2014年入选浙江省新世纪151高层次人才工程第二层次培养人员。主要研究方向是数理统计、面板数据计量经济分析、金融时间序列分析。近年来在《Journal of Econometrics》、《Test》、《Economic Modelling》、《Statistics and Probablity Letters》、《Communication in statistics-theory and methods》、《Linear Algebra and its Applications》等期刊发表学术论文20余篇。