Deep Inception Networks: A General End-to-End Framework for Multi-asset Quantitative Strategies

TL;DR

Deep Inception Networks (DINs) are a versatile end-to-end deep learning framework for systematic trading that automatically extracts features from time series and cross-sectional data, optimizing portfolio performance.

Contribution

This paper introduces DINs, a novel deep learning framework that automatically learns features for trading strategies and directly outputs portfolio positions, improving over prior methods.

Findings

DIN models outperform traditional benchmarks on futures data.

DINs are robust to transaction costs and perform consistently across seeds.

Attention and Variable Selection Networks enhance interpretability.

Abstract

We introduce Deep Inception Networks (DINs), a family of Deep Learning models that provide a general framework for end-to-end systematic trading strategies. DINs extract time series (TS) and cross sectional (CS) features directly from daily price returns. This removes the need for handcrafted features, and allows the model to learn from TS and CS information simultaneously. DINs benefit from a fully data-driven approach to feature extraction, whilst avoiding overfitting. Extending prior work on Deep Momentum Networks, DIN models directly output position sizes that optimise Sharpe ratio, but for the entire portfolio instead of individual assets. We propose a novel loss term to balance turnover regularisation against increased systemic risk from high correlation to the overall market. Using futures data, we show that DIN models outperform traditional TS and CS benchmarks, are robust to a range of transaction costs and perform consistently across random seeds. To balance the general nature of DIN models, we provide examples of how attention and Variable Selection Networks can aid the interpretability of investment decisions. These model-specific methods are particularly useful when the dimensionality of the input is high and variable importance fluctuates dynamically over time. Finally, we compare the performance of DIN models on other asset classes, and show how the space of potential features can be customised.