摘要：Large-dimensional factor models are drawing growing attention and widely applied to analyze the correlations of large datasets. Most related works focus on vector-valued data while nowadays matrix-valued or high-order tensor datasets are ubiquitous due to the accessibility to multiple data sources. In this talk, we review the matrix factor models and propose an iterative projection method for estimating the factor spaces. We show that the resulting convergence rate is faster than the typical rates that are conceivable from the literature on vector factor models as long as the observed data matrices are large. An easily satisfied sufficient condition on the projection direction to achieve the given rates for the projected estimators is provided. Moreover, we established the asymptotic distributions of the estimated row and column factor loadings. We also introduced an iterative approach to consistently determine the numbers of row and column factors. Two real data examples related to financial engineering and image recognition show that the projection estimators contribute to explaining portfolio variances and achieving accurate classification of handwritten digit numbers. Joint work with Yu Long (FDU), He Yong (USD), Zhang Xinsheng (FDU).
个人简介：孔新兵，现为南京审计大学教授、ISI elected member；主要研究兴趣为高频数据分析、髙维因子分析和经济金融计量分析；担任两个学术期刊编委；中国现场统计研究会多个分会常务理事；独立发表AoS、Biometrika成果三项；主持基金项目四项。