Contrastive Learning of Asset Embeddings from Financial Time Series

TL;DR

This paper introduces a contrastive learning framework for creating asset embeddings from financial time series, improving tasks like classification and portfolio optimization by capturing meaningful asset relationships.

Contribution

It proposes a novel contrastive learning approach tailored for financial data, utilizing statistical sampling and various loss functions to enhance asset representation quality.

Findings

Significantly outperforms existing baselines in industry classification.

Improves portfolio optimization results.

Effectively captures meaningful asset relationships.

Abstract

Representation learning has emerged as a powerful paradigm for extracting valuable latent features from complex, high-dimensional data. In financial domains, learning informative representations for assets can be used for tasks like sector classification, and risk management. However, the complex and stochastic nature of financial markets poses unique challenges. We propose a novel contrastive learning framework to generate asset embeddings from financial time series data. Our approach leverages the similarity of asset returns over many subwindows to generate informative positive and negative samples, using a statistical sampling strategy based on hypothesis testing to address the noisy nature of financial data. We explore various contrastive loss functions that capture the relationships between assets in different ways to learn a discriminative representation space. Experiments on real-world datasets demonstrate the effectiveness of the learned asset embeddings on benchmark industry classification and portfolio optimization tasks. In each case our novel approaches significantly outperform existing baselines highlighting the potential for contrastive learning to capture meaningful and actionable relationships in financial data.