A Generative Adversarial Graph Neural Network for Synthetic Time Series Data

2026-05-21

AI Summary

This work addresses the limitations of conventional generative models that rely on weak stationarity assumptions and struggle to capture the non-stationary characteristics inherent in financial time series. To overcome this, the authors propose Sig-Graph GAN, a novel framework that uniquely integrates path signatures, LSTM networks, and graph neural networks. By employing the visibility graph algorithm to transform temporal sequences into graph structures, the model jointly captures both autoregressive dynamics and geometric topological features within a generative adversarial framework. Evaluated on log-return data from multiple stock markets, Sig-Graph GAN significantly outperforms existing baselines, demonstrating its effectiveness in synthesizing high-fidelity, non-stationary financial time series.

Abstract

Generating synthetic data for financial time series poses challenges, especially considering their non-stationary nature. Traditional statistical time series models normally assume weak stationarity. However, this assumption can constrain their effectiveness. Deep learning models, particularly Generative Adversarial Networks (GANs), have exhibited considerable potential in emulating complex probability distributions. GANs employ a generator-discriminator framework, where the generator creates data samples, while the discriminator distinguishes real from generated data. In this research, we introduce the Sig-Graph GAN model, which integrates the time-series signature, offering a structured summary of its temporal evolution; the Long Short-Term Memory network, capturing its inherent autoregressive structure; and Graph Neural Networks (GNNs), leveraging geometric patterns within the time-series data. To employ GNNs optimally, we use the visibility graph algorithm to derive a graph-based representation of the underlying time series. Numerical evaluations demonstrate that the Sig-Graph GAN model outperforms baseline methods in replicating the distribution of logarithmic returns across different stock exchanges. The integration of the graph structure with the autoregressive component effectively captures both geometric and temporal patterns embedded in time-series data. This research advances the field of GAN models for time series by introducing a model capable of leveraging both autoregressive properties and geometric structures for synthetic data generation.

Problem

• synthetic time series
• non-stationarity
• financial time series
• generative modeling
• data generation

Innovation

• Generative Adversarial Networks
• Graph Neural Networks
• Time Series Signature
• Visibility Graph
• Non-stationary Time Series