Transfer Ranking in Finance: Applications to Cross-Sectional Momentum with Data Scarcity

Abstract

Modern cross-sectional strategies incorporating sophisticated neural architectures outperform traditional counterparts when applied to mature assets with long histories. However, deploying them on instruments with limited samples generally produces over-fitted models with degraded performance. In this paper, we introduce Fused Encoder Networks -- a hybrid parameter-sharing transfer ranking model which fuses information extracted using an encoder-attention module from a source dataset with a similar but separate module operating on a smaller target dataset of interest. This approach mitigates the issue of models with poor generalisability. Additionally, the self-attention mechanism enables interactions among instruments to be accounted for at the loss level during model training and inference time. We demonstrate the effectiveness of our approach by applying it to momentum strategies on the top ten cryptocurrencies by market capitalisation. Our model outperforms state-of-the-art benchmarks on most measures and significantly improves the Sharpe ratio. It continues to outperform baselines even after accounting for the high transaction costs associated with trading cryptocurrencies.

Keywords: Deep Learning, Transfer Learning, Information Retrieval, Learning to Rank, Neural Networks, Data Scarcity, Factor Investing, Cross-sectional Strategies, Portfolio Construction, Cryptocurrencies