Context-aware frequency-embedding networks for spatio-temporal portfolio selection

Liu, Ruirui, Huang, Huichou, Ruf, Johannes ORCID: 0000-0003-3616-2194 and Wu, Qingyao (2025) Context-aware frequency-embedding networks for spatio-temporal portfolio selection. In: SIAM International Conference on Data Mining, 2003-05-01 - 2003-05-03, CA., United States, USA. (In Press)

Text (CAFE_GCN_for_SIAM_DM (1)) - Accepted Version Download (2MB)

Abstract

Recent developments in the applications of deep reinforcement learning methods to portfolio selection have achieved superior performance to conventional methods. However, two major challenges remain unaddressed in these models and inevitably lead to the deterioration of model performance. First, asset characteristics often suffer from low and unstable signal-to-noise ratios, leading to poor learning robustness of the predictive feature representations. Second, the existing literature fails to consider the complexity and diversity in long-term and short-term spatio-temporal predictive relations between the feature sequences and portfolio objectives. To tackle these problems, we propose a novel Context-Aware Frequency-Embedding Graph Convolution Network (Cafe-GCN) for spatio-temporal portfolio selection. It contains three important modules: (1) frequencyembedding block that explicitly captures the short-term and long-term predictive information embedded in asset characteristics meanwhile filtering out the noise; (2) context-aware block that learns multiscale temporal dependencies in the feature space; and (3) multi-relation graph convolutional block that exploits both static and dynamic spatial relations among assets. Extensive experiments on two real-world datasets demonstrate that Cafe-GCN consistently outperforms proposed techniques in the literature.

Item Type:	Conference or Workshop Item (Paper)
Official URL:	https://www.siam.org/conferences-events/siam-confe...
Additional Information:	© 2025 The Author(s)
Divisions:	Mathematics
Subjects:	Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Date Deposited:	20 Jan 2025 11:21
Last Modified:	20 Jan 2025 11:33
URI:	http://eprints.lse.ac.uk/id/eprint/126929

Actions (login required)

View Item