Matrix-Factor-Augmented Regression

Xiong Cai, Xinbing Kong, Xinlei Wu and Peng Zhao

Journal of Business & Economic Statistics, 2025, vol. 43, issue 4, 1145-1157

Abstract: As matrix-variate observations are increasingly available, to incorporate the interplay between the multi-cross-sections, we introduce a matrix-factor-augmented regression model (M-FARM) that proposes to predict ahead of time with factors of matrix predictors augmented in the regression. We show that the estimation error in the factor matrices, estimated by the projection procedure in the first step, enters into the estimation error of the regression parameters and the prediction error of the response variable with an asymptotically negligible rate. The central limit theorems of the estimates of the regression parameters are established under some mild conditions. Forecasting intervals with a theoretical guarantee are given. Monte Carlo simulations justify the theoretical results. We find empirically that the augmented matrix factors do help in forecasting macroeconomic variables relative to the benchmark matrix autoregressive model and vector-factor-augmented regression model (V-FARM).

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/07350015.2025.2478986 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:jnlbes:v:43:y:2025:i:4:p:1145-1157

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/UBES20

DOI: 10.1080/07350015.2025.2478986

Access Statistics for this article

Journal of Business & Economic Statistics is currently edited by Eric Sampson, Rong Chen and Shakeeb Khan

More articles in Journal of Business & Economic Statistics from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().