Signature-Informed Transformer for Asset Allocation

2025/10/03 by Yoontae Hwang, Stefan Zohren, Hwang, Yoontae +1
Business, Management and Accounting · Computer Science · Decision Sciences · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #FOS: Economics and business #Financial Distress and Bankruptcy Prediction #Machine Learning (cs.LG) #Machine Learning in Healthcare #Portfolio Management (q-fin.PM) #Stock Market Forecasting Methods

paper · pdf · doi:10.48550/arxiv.2510.03129

openalex publication_date 2025/10/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Modern deep learning for asset allocation typically separates forecasting from optimization. We argue this creates a fundamental mismatch where minimizing prediction errors fails to yield robust portfolios. We propose the Signature Informed Transformer to address this by unifying feature extraction and decision making into a single policy. Our model employs path signatures to encode complex path dependencies and introduces a specialized attention mechanism that targets geometric asset relationships. By directly minimizing the Conditional Value at Risk we ensure the training objective aligns with financial goals. We prove that our attention module rigorously amplifies signature derived signals. Experiments across diverse equity universes show our approach significantly outperforms both traditional strategies and advanced forecasting baselines. The code is available at: https://anonymous.4open.science/r/Signature-Informed-Transformer-For-Asset-Allocation-DB88

Citations

• FactorGCL: A Hypergraph-Based Factor Model with Temporal Residual Contrastive Learning for Stock Returns Prediction
• Decision-informed Neural Networks with Large Language Model Integration for Portfolio Optimization
• Rough Transformers: Lightweight and Continuous Time Series Modelling through Signature Patching
• iTransformer: Inverted Transformers Are Effective for Time Series Forecasting
• A Time Series is Worth 64 Words: Long-term Forecasting with Transformers
• TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis
• Returns-Driven Macro Regimes and Characteristic Lead-Lag Behaviour between Asset Classes
• Signature Methods in Machine Learning
• Distributionally Robust End-to-End Portfolio Construction
• Non-stationary Transformers: Exploring the Stationarity in Time Series Forecasting
• Are Transformers Effective for Time Series Forecasting?
• ETSformer: Exponential Smoothing Transformers for Time-series Forecasting
• FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting
• HIST: A Graph-based Framework for Stock Trend Forecasting via Mining Concept-Oriented Shared Information
• Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting
• Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
• Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
• Sig-SDEs model for quantitative finance
• Neural Controlled Differential Equations for Irregular Time Series
• Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting
• Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting
• Deep Signature Transforms
• Deep Hedging
• Deep learning with long short-term memory networks for financial market predictions
• Attention Is All You Need
• A Primer on the Signature Method in Machine Learning
• Kernels for sequentially ordered data
• Extracting information from the signature of a financial data stream
• Differential Equations driven by Π-rough paths
• Learning to trade via direct reinforcement

Related

• Portfolio Transformer for Attention-Based Asset Allocation
• Asset Allocation: From Markowitz to Deep Reinforcement Learning
• Portfolio Optimization with 2D Relative-Attentional Gated Transformer
• Asset Pricing and Deep Learning
• Reinforcement-Learning Portfolio Allocation with Dynamic Embedding of Market Information