EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Modeling Stylized Facts in FX Markets with FINGAN-BiLSTM: A Deep Learning Approach to Financial Time Series

Dong-Jun Kim iD, Do-Hyeon Kim iD, Sun-Yong Choi iD

Entropy · 14.06.2025 · 27 (6) · 635

DOI: 10.3390/e27060635 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 14.06.2025. Verfasst von Dong-Jun Kim und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Finance and Big Data, Gachon University, Seongnam 13120, Republic of Korea. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 64 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT), Seoul R\&BD Program(QR240016) through the Seoul Business Agency(SBA) funded by Seoul Metropolitan Government., National Research Foundation of Korea (NRF).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We propose the financial generative adversarial network–bidirectional long short-term memory (FINGAN-BiLSTM) model to accurately reproduce the complex statistical properties and stylized facts, namely, heavy-tailed behavior, volatility clustering, and leverage effects observed in the log returns of the foreign exchange (FX) market. The proposed model integrates a bidirectional LSTM (BiLSTM) into the conventional FINGAN framework so that the generator, discriminator, and predictor networks simultaneously incorporate both past and future information, thereby overcoming the information loss inherent in unidirectional LSTM architectures. Experimental results, assessed using metrics such as the Kolmogorov–Smirnov statistic, demonstrate that FINGAN-BiLSTM effectively mimics the distributional and dynamic patterns of actual FX data. In particular, the model significantly reduces the maximum cumulative distribution discrepancy in assets with high standard deviations and extreme values, such as the Canadian dollar (CAD) and the Mexican Peso (MXN), while precisely replicating dynamic features like volatility clustering and leverage effects, thereby outperforming conventional models. The findings suggest that the proposed deep learning–based forecasting model holds significant promise for practical applications in financial risk assessment, derivative pricing, and portfolio optimization, and they highlight the need for further research to enhance its generalization capabilities through the integration of exogenous economic variables.

Schlagwörter

Stichwörter der Autor:innen

BiLSTMFINGANforecastingforeign exchangestylized fact

Autor:innen und Institutionen

  • Department of Finance and Big Data, Gachon University, Seongnam 13120, Republic of Korea.

ORCID-Profile: Dong-Jun Kim, Do-Hyeon Kim, Sun-Yong Choi

Förderung

  • National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT)
  • Seoul R\&BD Program(QR240016) through the Seoul Business Agency(SBA) funded by Seoul Metropolitan Government.
  • National Research Foundation of Korea (NRF)
  • Seoul R&BD Program

Zitieren

APA

Kim, D. J., Kim, D. H., & Choi, S. Y. (2025). Modeling Stylized Facts in FX Markets with FINGAN-BiLSTM: A Deep Learning Approach to Financial Time Series. Entropy, 27(6), 635. https://doi.org/10.3390/e27060635

Vancouver

Kim DJ, Kim DH, 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

BibTeX
@article{Kim2025,
  author = {Kim, Dong-Jun and Kim, Do-Hyeon and Choi, Sun-Yong},
  title = {Modeling Stylized Facts in FX Markets with FINGAN-BiLSTM: A Deep Learning Approach to Financial Time Series},
  journal = {Entropy},
  year = {2025},
  volume = {27},
  number = {6},
  pages = {635},
  doi = {10.3390/e27060635}
}
Kostenlos starten →