Transfer Learning in High-Dimensional Conditional Factor Models
Prof. Yubo TAO
Assistant Professor of Economics
Department of Economics
FSS, UM
Date: 11/11/2025 (Tuesday)
Time: 13:00 to 14:00
Venue: FBA Lobby
Abstract
This paper considers the estimation and prediction of a high-dimensional conditional factor model in the setting of transfer learning, where, in addition to observations from the target model, auxiliary datasets are available. To effectively utilize the auxiliary datasets, we employ a transfer learning algorithm in conjunction with the instrumented principal component analysis (IPCA) to estimate the conditional factor models. Given the informativeness of the auxiliary datasets, we propose a trans-IPCA algorithm and derive its estimation error bounds. We prove that when the target and sources are sufficiently close to each other, these bounds can be improved over those of the classical IPCA estimator and its penalized variants, using only target data, under mild conditions. When the set of informative auxiliary data is unknown, we introduce a data-driven and algorithm-free procedure to detect transferable samples. Monte Carlo simulations confirm the superior performance of the proposed estimator compared to classical and penalized IPCA models, both in-sample and out-of-sample.
Speaker
Yubo Tao is an Assistant Professor of Economics in the Faculty of Social Sciences, University of Macau. He obtained his PhD in Economics from Singapore Management University. His research fields are econometric theory, financial econometrics, and empirical asset pricing. His work has been published in numerous leading finance and economics journals, including Management Science, Journal of Econometrics, and Journal of Business & Economic Statistics. His research has been funded by the NSFC grant and he currently serves as an associate editor for the International Journal of Finance & Economics and the Journal of Economic Surveys.
All are welcome!
