报告题目：Multiple quasi-maximum likelihood estimation of break points in

high-dimensional factor models

报告人：段江涛

时间：11月20日（周三）下午16:00-18:00

地点：西安交通大学创新港涵英楼8121会议室

报告人简介：

段江涛，现为西安电子科技大学数学与统计学院统计系菁英副教授。2021年获东北师范大学统计学博士学位。曾赴美国哥伦比亚大学、香港城市大学进行访问与合作研究。研究兴趣为高维因子模型，面板数据，试验设计。主持和参与国家自然科学基金项目和国家社会科学基金项目多项，多篇文章发表于Journal of Econometrics, Canadian Journal of Statistics, Journal of Statistical Computation and Simulation, British Journal of Mathematical and Statistical Psychology.

摘要：

This paper explores the estimation of high-dimensional factor models where the factor loadings undergo an unknown number of structural changes over time. Given that a model with multiple changes in factor loadings can be observationally indistinguishable from one with constant loadings but varying factor variances, this reduces the high-dimensional structural change problem to a lower-dimensional one. We propose a quasi-maximum likelihood (QML) method for detecting breakpoints, establishing the convergence rate between the eigenvalues of the estimated factor covariance matrix and those of the sub-regime covariance matrix, and demonstrating that the QML estimators are consistent when the number of factors in the combined regime exceeds the minimum number of factors in the two separate regimes. Furthermore, we show that the distance between the QML estimated and the true break points is bounded when the number of factors in the combined regime equals that in both separate regimes. We also introduce an information criterion (IC) for estimating the number of breakpoints and show that, with probability approaching one, our method accurately identifies the correct number of structural changes. Simulation studies reveal strong finite-sample performance. Applying this methodology to the FRED-MD dataset, we detect five structural breaks in the factor loadings between 1959 and 2024.

经济与金融学院

2024年11月18日