Generative Adversarial Networks (GANs) for Data Augmentation in Emerging Financial Market Prediction: A Systematic Literature Review

Authors

DOI:

https://doi.org/10.61838/jmbs.495

Keywords:

Generative Adversarial Networks, GAN, data augmentation, emerging financial markets, financial forecasting, synthetic data, time, series prediction, deep learning

Abstract

This systematic literature review aimed to synthesize empirical evidence on the effectiveness, methodological characteristics, and predictive contribution of Generative Adversarial Network-based data augmentation in emerging financial market prediction. A systematic review was conducted in accordance with PRISMA 2020 principles using Scopus, Web of Science Core Collection, IEEE Xplore, ScienceDirect, and SpringerLink. Studies published from January 2014 to September 30, 2026 were screened according to predefined eligibility criteria. The initial search identified 1,284 records; after removal of 286 duplicates, 998 titles and abstracts were screened, 142 full-text articles were assessed, and 33 empirical studies were ultimately included. Data were extracted on market type, GAN architecture, augmentation strategy, downstream prediction model, validation procedure, synthetic-data quality, and predictive performance. Methodological quality was assessed using a 12-criterion framework, and findings were synthesized descriptively and thematically because heterogeneity precluded formal meta-analysis. Among the 33 included studies, 27 (81.8%) reported clear improvement in the primary predictive outcome following GAN-based augmentation, whereas six (18.2%) reported mixed or negligible benefits. TimeGAN-type models, conditional GANs, and WGAN-GP architectures produced the most consistent improvements, while standard and highly complex hybrid GANs showed greater variability. In regression tasks, RMSE reductions ranged from 5.8% to 21.6% and MAE reductions from 4.3% to 18.9%. Directional-classification studies reported accuracy gains of 2.7–9.8 percentage points and F1-score improvements of 3.1–11.4 points. The strongest benefits occurred in small-sample, imbalanced, rare-event, and temporally complex settings. However, methodological weaknesses remained in synthetic-data fidelity assessment, robustness testing, time-aware validation, and computational reproducibility. GAN-based data augmentation appears to enhance predictive performance in emerging financial markets, particularly when generative architectures are matched to temporal structure, class imbalance, and tail-risk characteristics, although gains depend strongly on rigorous validation and realistic synthetic-data generation.

Downloads

Download data is not yet available.

References

1. Casolaro A, Capone V, Iannuzzo G, Camastra F. Deep Learning for Time Series Forecasting: Advances and Open Problems. Information. 2023;14(11):598. doi: 10.3390/info14110598.

2. Henrici A, Osterrieder J. Editorial: Artificial Intelligence in Finance and Industry: Highlights From 6 European COST Conferences. Frontiers in Artificial Intelligence. 2022;5. doi: 10.3389/frai.2022.1007074.

3. Kumar U, Sinha B. The Advent of Generative AI and Financial Industry. International Journal of Finance. 2024;9(5):55-70. doi: 10.47941/ijf.2210.

4. Lee DKC, Guan C, Yu Y, Ding Q. A Comprehensive Review of Generative AI in Finance. 2024. doi: 10.20944/preprints202407.2109.v1.

5. Zhang Y, Jiang Z, Peng C, Zhu X, Wang G. Management Analysis Method of Multivariate Time Series Anomaly Detection in Financial Risk Assessment. Journal of Organizational and End User Computing. 2024;36(1):1-19. doi: 10.4018/joeuc.342094.

6. Gong Y, Hu P, Zhang Z, Liu P, Li Z, Zhang R, et al. Data-Centric Generative and Adaptive Detection Framework for Abnormal Transaction Prediction. Electronics. 2026;15(3):633. doi: 10.3390/electronics15030633.

7. Gao X. A Two-Stage Deep Learning Model for Risk Identification in Green Supply Chain Finance. Scientific Reports. 2026;16(1). doi: 10.1038/s41598-026-46830-1.

8. Xi J. Multimodal Conflict-Aware and Generative-Enhanced AI for Early Startup Survival and Risk Prediction. 2025. doi: 10.21203/rs.3.rs-8365925/v1.

9. Liu D, Wu Y, Hong D, Wang S. Time Series Data Augmentation Method of Small Sample Based on Optimized Generative Adversarial Network. Concurrency and Computation Practice and Experience. 2022;34(27). doi: 10.1002/cpe.7331.

10. Hazra D, Shafqat W, Byun Y-C. Generating Synthetic Data to Reduce Prediction Error of Energy Consumption. Computers Materials & Continua. 2022;70(2):3151-67. doi: 10.32604/cmc.2022.020143.

11. Zhong S, Wang X, Xu B, Wu H, Ding M. Hybrid Network Model Based on Data Enhancement for Short-Term Power Prediction of New PV Plants. Journal of Modern Power Systems and Clean Energy. 2024;12(1):77-88. doi: 10.35833/mpce.2022.000759.

12. Yilmaz B. Generative Adversarial Network for Load Data Generation: Türkiye Energy Market Case. Mathematical Modelling and Numerical Simulation With Applications. 2023;3(2):141-58. doi: 10.53391/mmnsa.1320914.

13. Naveed MH, Hashmi U, Tajved N, Sultan N. Is Synthetic the New Real? Performance Analysis of Time Series Generation Techniques With Focus on Network Load Forecasting. 2021. doi: 10.36227/techrxiv.17296235.v1.

14. Dogariu M, Stefan LD, Boteanu B, Lamba C, Kim B, Ionescu BE. Generation of Realistic Synthetic Financial Time-Series. Acm Transactions on Multimedia Computing Communications and Applications. 2022;18(4):1-27. doi: 10.1145/3501305.

15. Ramzan F, Sartori C, Consoli S, Recupero DR. Generative Adversarial Networks for Synthetic Data Generation in Finance: Evaluating Statistical Similarities and Quality Assessment. Ai. 2024;5(2):667-85. doi: 10.3390/ai5020035.

16. Bai X, Zhuang S, Xie H, Guo L. Leveraging Generative Artificial Intelligence for Financial Market Trading Data Management and Prediction. 2024. doi: 10.20944/preprints202407.0084.v1.

17. Gándara ÁR, Abril LG. Generative Adversarial Networks in Business and Social Science. 2024. doi: 10.20944/preprints202407.2122.v1.

18. Fu W, Hirsa A. Simulating Financial Time Series Using Attention. 2022. doi: 10.48550/arxiv.2207.00493.

19. Pardo FdM, Schwendner P, Wunsch M. Tackling the Exponential Scaling of Signature-Based Generative Adversarial Networks for High-Dimensional Financial Time-Series Generation. The Journal of Financial Data Science. 2022;4(4):110-32. doi: 10.3905/jfds.2022.1.109.

20. Hirano M, Sakaji H, Izumi K. Policy Gradient Stock Gan for Realistic Discrete Order Data Generation in Financial Markets. 2022. doi: 10.2139/ssrn.4095304.

21. Xia H, Sun S, Wang X, An B. Market-Gan: Adding Control to Financial Market Data Generation With Semantic Context. Proceedings of the Aaai Conference on Artificial Intelligence. 2024;38(14):15996-6004. doi: 10.1609/aaai.v38i14.29531.

22. Abdelhady AS, Dahmani N, Abouelmagd LM, Darwish A, Hassanien AE. Green Finance Growth Prediction Model Based on Time-Series Conditional Generative Adversarial Networks. Plos One. 2024;19(7):e0306874. doi: 10.1371/journal.pone.0306874.

23. Podobiński A, Chudziak JA. Financial Time Series Augmentation Using Transformer Based GAN Architecture. 2026. doi: 10.48550/arxiv.2602.17865.

24. Gadhi AHA, Peiris S, Allen DE. Improving Volatility Forecasting: A Study Through Hybrid Deep Learning Methods With WGAN. Journal of Risk and Financial Management. 2024;17(9):380. doi: 10.3390/jrfm17090380.

25. Hu Z, Yang X, Jiang W. Research on Deep Learning Financial Volatility Prediction Method Based on Signal Decomposition and Data Augmentation. 2025. doi: 10.21203/rs.3.rs-7726466/v1.

26. Kim D-J, Kim D-H, Choi SY. Modeling Stylized Facts in FX Markets With FINGAN-BiLSTM: A Deep Learning Approach to Financial Time Series. Entropy. 2025;27(6):635. doi: 10.3390/e27060635.

27. Labiad B, Berrado A, Benabbou L. Predicting Extreme Events in the Stock Market Using Generative Adversarial Networks. International Journal of Advances in Intelligent Informatics. 2023;9(2):218. doi: 10.26555/ijain.v9i2.898.

28. Chandra R. Extreme Value Forecasting Using Relevance-Based Data Augmentation With Deep Learning Models. 2025. doi: 10.48550/arxiv.2510.02407.

29. Hirnschall D. Semi-Supervised Bayesian GANs With Log-Signatures for Uncertainty-Aware Credit Card Fraud Detection. Mathematics. 2025;13(19):3229. doi: 10.3390/math13193229.

30. Sundaravadivel P, R AI. Federated Deep Learning and Explainable AI for Real-Time Credit Card Fraud Detection in Highly Imbalanced Transaction Streams. 2025. doi: 10.21203/rs.3.rs-7587282/v1.

31. Hellermann J, Qian Q, Shah A. Financial Time Series Data Augmentation With Generative Adversarial Networks and Extended Intertemporal Return Plots. 2022. doi: 10.48550/arxiv.2205.08924.

32. Prenestino FB, Barbierato E, Gatti A. Robust Synthetic Data Generation for Sequential Financial Models Using Hybrid Variational Autoencoder–Markov Chain Monte Carlo Architectures. Future Internet. 2025;17(2):95. doi: 10.3390/fi17020095.

33. Mahajan RA, Dey R, Khan M. Enhancing Financial Predictive Modeling With Synthetic Data Using Generative Approach. 2025. doi: 10.21203/rs.3.rs-6178330/v1.

34. Kalina B, Lee J-H, Na K. Enhancing Portfolio Performance Through Financial Time-Series Decomposition-Based Variational Encoder-Decoder Data Augmentation. Symmetry. 2024;16(3):283. doi: 10.3390/sym16030283.

35. Waller J, Caruso F, Makris D, Nilavalan R, Liang X. Quantum Generative Diffusion Model for Real-World Time Series. 2026. doi: 10.48550/arxiv.2606.27561.

36. Shi X, Hu P. A GAN-Based Framework for Robust Data Synthesis in Satellite Internet Observations. 2026. doi: 10.48550/arxiv.2607.24790.

Downloads

Publication Timeline

Published
Submitted
Revised
Accepted

How to Cite

Forough Nejad, H. ., Ghavami, B., & Sadri, M. . (2026). Generative Adversarial Networks (GANs) for Data Augmentation in Emerging Financial Market Prediction: A Systematic Literature Review. Journal of Management and Business Solutions, 4(2), 1-23. https://doi.org/10.61838/jmbs.495