Transfer learning is an emerging and popular paradigm for utilizing existing knowledge from previous learning tasks to improve the performance of new ones. In this paper, we propose a novel concept of transfer risk and and analyze its properties to evaluate transferability of transfer learning. We apply transfer learning techniques and this concept of transfer risk to stock return prediction and portfolio optimization problems. Numerical results demonstrate a strong correlation between transfer risk and overall transfer learning performance, where transfer risk provides a computationally efficient way to identify appropriate source tasks in transfer learning, including cross-continent, cross-sector, and cross-frequency transfer for portfolio optimization.
@article{arxiv.2311.03283,
title = {Risk of Transfer Learning and its Applications in Finance},
author = {Haoyang Cao and Haotian Gu and Xin Guo and Mathieu Rosenbaum},
journal= {arXiv preprint arXiv:2311.03283},
year = {2023}
}
Comments
arXiv admin note: substantial text overlap with arXiv:2307.13546