arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2512.21791cs.LGq-fin.CP
收藏

合成金融数据生成以提升金融建模

Synthetic Financial Data Generation for Enhanced Financial Modelling

AIMS Rwanda(非洲数学科学研究所(Rwanda));African Institute for Mathematical Sciences(非洲数学科学研究所);College of Engineering, Carnegie Mellon University(卡内基梅隆大学工程学院);Sefako Makgatho Health Sciences University (SMU)(塞法科·马加索健康科学大学);AI Research and Innovation Nexus for Africa (AIRINA Labs)(非洲人工智能研究与创新 nexus(AIRINA 实验室));AI.Technipreneurs, Bénin(AI.Technipreneurs(贝宁))

Christophe D. Hounwanou, Yae Ulrich Gaba, Pierre Ntakirutimana

AI总结:

本文提出统一评估框架,比较了ARIMA-GARCH、VAEs和TimeGAN在生成合成金融数据中的性能,发现TimeGAN在现实感和时间一致性方面表现最佳。

摘要:

金融领域中的数据稀缺性和保密性常常阻碍模型开发和稳健测试。本文提出了一种统一的多标准评估框架用于合成金融数据,并将其应用于三种具有代表性的生成范式:统计ARIMA-GARCH基线、变分自编码器(VAEs)和时间序列生成对抗网络(TimeGAN)。使用历史S&P 500每日数据,我们评估了保真度(最大均值差异,MMD)、时间结构(自相关性和波动性聚类)以及在下游任务中的实用价值,特别是均值-方差投资组合优化和波动率预测。实证结果表明,ARIMA-GARCH能够捕捉线性趋势和条件波动性,但无法再现非线性动态;VAEs生成平滑轨迹但低估了极端事件;而TimeGAN在现实感和时间一致性之间达到了最佳平衡(例如,TimeGAN在5个种子下的MMD最低为1.84e-3)。最后,我们提出了根据应用需求和计算约束选择生成模型的实用指南。我们的统一评估协议和可重复的代码库旨在标准化合成金融数据研究中的基准测试。

英文摘要:

Data scarcity and confidentiality in finance often impede model development and robust testing. This paper presents a unified multi-criteria evaluation framework for synthetic financial data and applies it to three representative generative paradigms: the statistical ARIMA-GARCH baseline, Variational Autoencoders (VAEs), and Time-series Generative Adversarial Networks (TimeGAN). Using historical S and P 500 daily data, we evaluate fidelity (Maximum Mean Discrepancy, MMD), temporal structure (autocorrelation and volatility clustering), and practical utility in downstream tasks, specifically mean-variance portfolio optimization and volatility forecasting. Empirical results indicate that ARIMA-GARCH captures linear trends and conditional volatility but fails to reproduce nonlinear dynamics; VAEs produce smooth trajectories that underestimate extreme events; and TimeGAN achieves the best trade-off between realism and temporal coherence (e.g., TimeGAN attained the lowest MMD: 1.84e-3, average over 5 seeds). Finally, we articulate practical guidelines for selecting generative models according to application needs and computational constraints. Our unified evaluation protocol and reproducible codebase aim to standardize benchmarking in synthetic financial data research.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑