What type of study is this?

This is a Quantitative Study study.

October 12, 2025

Robust factor analysis for high-dimensional data with outliers

PZPing ZhouCentral South University JLJialiang LiQingdao University of Science and Technology XGXiao GuoUniversity of Science and Technology of China

Key Points

Proposed new methods significantly enhance factor analysis performance in the presence of outliers, and show greater consistency.
The proposed methods outperform existing approaches, indicating improved robustness in analyzing high-dimensional data under contamination.
The study establishes consistency for new estimators under mild conditions, with numerical examples supporting the findings.
A new class of factor model incorporating a sparse outlier matrix facilitates more accurate estimation in challenging data settings.

Abstract

Outliers are commonly encountered in financial and economic data, and pose significant challenges to the study of the conventional factor models. This paper focuses on factor analysis for high-dimensional data in the presence of random outliers. We first propose a new class of factor model which incorporates a sparse outlier matrix. We then develop a class of robust methods to estimate the factor matrix and the number of factors. Under mild conditions, consistency of the proposed estimators is established. Judging from the numerical examples, we observe that the proposed estimation methods outperform existing approaches. Our methods significantly enhance the performance of the high-dimensional factor analysis when the data are contaminated by outliers.

AI से पूछें

Bookmark

View Full Paper