실천공학교육논문지 (Journal of Practical Engineering Education)

한국실천공학교육학회 (Korean Institute for Pratical Engineering Education)
권호 표지
DOI QR코드

DOI QR Code

심층학습을 이용한 한국종합주가지수의 특성분석

Characteristic Analysis of Kospi Index Using Deep Learning

  • 한상일 (한국기술교육대학교 산업경영학과)
  • Snag-Il Han (Dep. of Industrial Management, Korea University of Technology and Education)
  • 투고 : 2024.01.16
  • 심사 : 2024.01.31
  • 발행 : 2024.02.29
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본고는 Kospi와 S&P500 지수를 이용해 한미 주식시장 간 차이를 보고 이를 통해 정책적 시사점을 논하고자 한다. 이를 위해 기존 시계열 분석 방법에 더해 심층학습 방법으로 시장간 비교를 하되 주가 예측력, 자료 생성 능력 측면에서 비교를 했다. 월별자료에서 시계열간 차이는 크지 않고 일별 자료에서 안정성 측면에서 차이가 약하며, 예측력이나 모의자료 생성에서도 차이가 크지 않았다. 본 연구결과와 같이 시장가격 움직임의 패턴이 한미간에 차이가 크지 않다면, 공매도의 부작용에 대한 대책으로 담보비율, 보고주기와 같은 직접적 규제보다 미국과 유사하게 투자자들의 자산운용 전략에 영향을 미치는 장기 주식보유에 대한 세제혜택과 같은 제도개편이 효과적이라 본다.

This paper examines the differences between the Korean and American stock markets using the Kospi and S&P 500 indices and discusses policy implications through them. To this end, in addition to the existing time series analysis method, a deep learning method was used to compare markets, and the comparison was made in terms of stock price forecasting ability and data generation ability. In monthly data, the difference between time series was not large, and in daily data, the difference in terms of stability was weak, and there was no significant difference in predictive power or simulation data generation. As shown in the results of this study, if there is not much difference in market price movement patterns between Korea and the United States, tax benefits for long-term stocks investment will be effective against the side effects of short selling.

키워드

참고문헌

  1. J. Campbell, A. Lo, and A. MacKinlay, The Econometrics of Financial Markets, 1997.
  2. J. Cochrane, Asset Pricing: Revised Edition, Princeton, 2005.
  3. S. Chun, "Predicting Korean stock market return with Financial and macro variables," The Journal of Insurance and Finance, vol. 31, no. 1, pp. 87-113, 2020. https://doi.org/10.23842/jif.2020.31.1.003
  4. J. I. Yoon, "The evaluations and the international comparisons of the integration with U.S. stock market," Journal of Money & Finance, vol. 21, no. 1, pp. 55-92.
  5. H. Choi, "Stock prediction analysis through artificial intelligence using big data," Journal of the Korea Institute of Information and Communication Engineering, vol. 25, no. 10, pp. 1435-1440, 2021.
  6. S. H. Hong, "Cryptocurrency automatic trading research by using facebook deep learning algorithm," Journal of Digital Convergence, vol. 19, no. 11, pp. 359-364, 2021.
  7. J. Jung and J. Kim, "A performance analysis by adjusting learning methods in stock price prediction model using LSTM," Journal of Digital Convergence, vol. 18, no. 11, pp. 259-266, 2020. https://doi.org/10.14400/JDC.2020.18.11.259
  8. J. Yoo, Y. Soun, Y. Park, and U. Kang, "Accurate multivariate stock movement prediction via data-axis transformer with multi-level contexts," ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) 2021.
  9. J. Y. Lee, J. H. Lee, B. G. Choi, and J. W. Song, "Trading algorithm selection using Time-series generative adversal networks," Smart Media Journal, vol. 11, no. 1, pp. 38- 45, 2022. https://doi.org/10.30693/SMJ.2022.11.1.38
  10. D. Kobiela, D. Krefta, W. Krol, and P. Weichbroth, "ARIMA vs LSTM on NASDAQ stock exchange data," Procedia Computer Science, vol. 207, pp. 3836-3845, 2022. https://doi.org/10.1016/j.procs.2022.09.445
  11. H. Lin, C. Chen, A. Jafari, and G. Huang, "Stock price prediction using generative adversarial networks," Journal of Computer Science, vol. 17, no. 3, pp. 188-196, 2021. https://doi.org/10.3844/jcssp.2021.188.196
  12. N. Foumani, L. Miller, C. Tan, G. Webb, G. Forestier, and M. Salehi, "Deep learning for time series classification and extrinsic regression: A current survey," arXiv: 2302.02515, 2023.
  13. S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," Neural Computation, vol. 9, no. 8, 1735-1780, 1997. https://doi.org/10.1162/neco.1997.9.8.1735
  14. H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, and W. Zhang, "Informer: Beyond efficient transformer for long sequence time-series forecasting," arXiv 2012.07436, 2021.
  15. I. Goodfellow, P. Jean, M. Mehdi, X. Bing, W. David, S. Ozair, C. Aaron, and Y. Bengio, "Generative adversarial nets," NIPS, pp. 2672-2680, 2014.
  16. J. Yoon and D. Jarrett, "Time-series generative adversarial networks," Neural Information Processing Systems (NeurIPS), 2019.