Quantum generative modeling for financial time series with temporal correlations

2025-07-29

AI Summary

Traditional GANs struggle to simultaneously capture authentic statistical distributions and temporal dependency structures in financial time series generation. To address this, we propose a quantum generative adversarial network (QGAN) framework: a quantum generator models intricate temporal correlations, while a classical discriminator enforces distributional alignment. Leveraging both full quantum circuits and tensor network approximations, we systematically investigate the impact of hyperparameters—including circuit depth and bond dimension—on generation quality. Experiments demonstrate that our method faithfully reproduces target distributions and multiscale temporal dependencies (e.g., autocorrelation and volatility clustering), significantly enhancing data augmentation performance under limited historical data—outperforming classical GANs and Wasserstein-GAN. The core contribution is the first integration of quantum correlation mechanisms into time-series generation, establishing an interpretable and controllable quantum-enhanced paradigm for financial data synthesis.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation

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
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Quantum generative adversarial networks (QGANs) have been investigated as a method for generating synthetic data with the goal of augmenting training data sets for neural networks. This is especially relevant for financial time series, since we only ever observe one realization of the process, namely the historical evolution of the market, which is further limited by data availability and the age of the market. However, for classical generative adversarial networks it has been shown that generated data may (often) not exhibit desired properties (also called stylized facts), such as matching a certain distribution or showing specific temporal correlations. Here, we investigate whether quantum correlations in quantum inspired models of QGANs can help in the generation of financial time series. We train QGANs, composed of a quantum generator and a classical discriminator, and investigate two approaches for simulating the quantum generator: a full simulation of the quantum circuits, and an approximate simulation using tensor network methods. We tested how the choice of hyperparameters, such as the circuit depth and bond dimensions, influenced the quality of the generated time series. The QGAN that we trained generate synthetic financial time series that not only match the target distribution but also exhibit the desired temporal correlations, with the quality of each property depending on the hyperparameters and simulation method.

Problem

• Generate synthetic financial time series with quantum GANs
• Ensure synthetic data exhibits desired temporal correlations
• Optimize hyperparameters for distribution and correlation quality

Innovation

• Quantum GANs for financial time series
• Hybrid quantum-classical generator-discriminator approach
• Tensor network simulation for quantum circuits