Deep Learning for Financial Time Series: A Large-Scale Benchmark of Risk-Adjusted Performance

2026-03-02

AI Summary

This study addresses the optimization of the Sharpe ratio in financial time series forecasting and position management, explicitly accounting for downside risk, transaction costs, and model robustness. Leveraging a large-scale real-world dataset of cross-asset futures, it presents the first systematic out-of-sample evaluation of diverse model architectures—including linear models, RNNs, Transformers, state-space models, and hybrid designs such as VSN-LSTM, VSN-xLSTM, and LSTM-PatchTST—with the Sharpe ratio as the central objective. The results demonstrate that VSN-LSTM achieves the highest Sharpe ratio, while VSN-xLSTM and LSTM-PatchTST exhibit superior downside risk control. Notably, xLSTM proves most robust to trading frictions. This work establishes a comprehensive benchmark and offers practical guidance for applying time series modeling in quantitative investment.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

We present a large scale benchmark of modern deep learning architectures for a financial time series prediction and position sizing task, with a primary focus on Sharpe ratio optimization. Evaluating linear models, recurrent networks, transformer based architectures, state space models, and recent sequence representation approaches, we assess out of sample performance on a daily futures dataset spanning commodities, equity indices, bonds, and FX spanning 2010 to 2025. Our evaluation goes beyond average returns and includes statistical significance, downside and tail risk measures, breakeven transaction cost analysis, robustness to random seed selection, and computational efficiency. We find that models explicitly designed to learn rich temporal representations consistently outperform linear benchmarks and generic deep learning models, which often lead the ranking in standard time series benchmarks. Hybrid models such as VSN with LSTM, a combination of Variable Selection Networks (VSN) and LSTMs, achieves the highest overall Sharpe ratio, while VSN with xLSTM and LSTM with PatchTST exhibit superior downside adjusted characteristics. xLSTM demonstrates the largest breakeven transaction cost buffer, indicating improved robustness to trading frictions.

Problem

• financial time series
• risk-adjusted performance
• Sharpe ratio
• deep learning
• position sizing

Innovation

• financial time series
• Sharpe ratio optimization
• deep learning benchmark
• risk-adjusted performance
• hybrid sequence models