Beyond Visual Realism: Toward Reliable Financial Time Series Generation

2026-01-19

AI Summary

This work proposes a novel generative adversarial network (GAN) framework for financial time series synthesis that explicitly preserves key stylized facts—such as fat tails, volatility clustering, and return asymmetry—which are critical for trading performance yet often overlooked by existing models despite their visual realism. By reformulating these empirical regularities as differentiable structural constraints and jointly optimizing them with the adversarial loss, the method ensures that generated sequences closely match real market data not only in statistical properties but also in backtested trading outcomes. Experiments on the Shanghai Composite Index (2004–2024) demonstrate that the proposed approach effectively maintains these stylized facts and significantly enhances the robustness of momentum strategy backtests, outperforming current state-of-the-art baselines.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Generative models for financial time series often create data that look realistic and even reproduce stylized facts such as fat tails or volatility clustering. However, these apparent successes break down under trading backtests: models like GANs or WGAN-GP frequently collapse, yielding extreme and unrealistic results that make the synthetic data unusable in practice. We identify the root cause in the neglect of financial asymmetry and rare tail events, which strongly affect market risk but are often overlooked by objectives focusing on distribution matching. To address this, we introduce the Stylized Facts Alignment GAN (SFAG), which converts key stylized facts into differentiable structural constraints and jointly optimizes them with adversarial loss. This multi-constraint design ensures that generated series remain aligned with market dynamics not only in plots but also in backtesting. Experiments on the Shanghai Composite Index (2004--2024) show that while baseline GANs produce unstable and implausible trading outcomes, SFAG generates synthetic data that preserve stylized facts and support robust momentum strategy performance. Our results highlight that structure-preserving objectives are essential to bridge the gap between superficial realism and practical usability in financial generative modeling.

Problem

• financial time series generation
• stylized facts
• tail events
• trading backtest
• generative models

Innovation

• Stylized Facts
• Financial Time Series Generation
• GAN
• Structural Constraints
• Backtesting