The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Factor augmentation for cryptocurrency return forecasting

암호화폐 수익률 예측력 향상을 위한 요인 강화

  • Yeom, Yebin (Department of Statistics, Sungkyunkwan University) ;
  • Han, Yoojin (Department of Economics, Sungkyunkwan University) ;
  • Lee, Jaehyun (Department of Mathematics, Sungkyunkwan University) ;
  • Park, Seryeong (Department of Statistics, Sungkyunkwan University) ;
  • Lee, Jungwoo (Department of Statistics, Sungkyunkwan University) ;
  • Baek, Changryong (Department of Statistics, Sungkyunkwan University)
  • 염예빈 (성균관대학교 통계학과) ;
  • 한유진 (성균관대학교 경제학과) ;
  • 이재현 (성균관대학교 수학과) ;
  • 박세령 (성균관대학교 통계학과) ;
  • 이정우 (성균관대학교 통계학과) ;
  • 백창룡 (성균관대학교 통계학과)
  • Received : 2021.10.01
  • Accepted : 2021.12.06
  • Published : 2022.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, we propose factor augmentation to improve forecasting power of cryptocurrency return. We consider financial and economic variables as well as psychological aspect for possible factors. To be more specific, financial and economic factors are obtained by applying principal factor analysis. Psychological factor is summarized by news sentiment analysis. We also visualize such factors through impulse response analysis. In the modeling perspective, we consider ARIMAX as the classical model, and random forest and deep learning to accommodate nonlinear features. As a result, we show that factor augmentation reduces prediction error and the GRU performed the best amongst all models considered.

본 연구는 외부 요인을 모형에 강화시켜 암호화폐 수익률 예측력을 향상시키는 방법에 대해서 다루고 있다. 고려한 요인으로는 크게 나누어 금융 경제적 요인 및 심리적 요인을 고려하였다. 먼저 금융 경제적 요인을 반용하기 위해서 주성분 요인을 사용하여 수 많은 변수를 차원축소를 통해서 모형에 반영하였다. 또한 심리적 요인을 위해서는 뉴스 기사 데이터를 활용하여 산출해낸 감성지수를 활용하였다. 이러한 요인들은 충격반응함수 분석을 통해서 요인들의 의미와 영향력을 시각화하였다. 또한 전통적인 ARIMAX 뿐 만 아니라 랜덤포레스트 및 딥러닝 모형을 활용하여 비선형성을 반영하였다. 그 결과 요인 강화가 암호화폐 수익률 예측력을 향상시킴을 실증분석을 통해 밝혔으며 그 중에서 딥러닝 모형인 GRU가 가장 좋은 예측 성능을 보임을 관찰하였다.

Keywords

References

  1. Bai J and Ng S (2008). Large Dimensional Factor Analysis, Now Publishers Inc.
  2. Bariviera AF, Basgall MJ, Hasperue W, and Naiouf M (2017). Some stylized facts of the Bitcoin market, Physica A: Statistical Mechanics and its Applications, 484, 82-90. https://doi.org/10.1016/j.physa.2017.04.159
  3. Bernanke BS, Boivin J, and Eliasz P (2005). Measuring the effects of monetary policy: a factor-augmented vector autoregressive (favar) approach, The Quarterly Journal of Economics, 120, 387-422. https://doi.org/10.1162/0033553053327452
  4. Biau G and Scornet E (2016). A random forest guided tour, Test, 25, 197-227. https://doi.org/10.1007/s11749-016-0481-7
  5. Cho K, Van Merrienboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, and Bengio Y (2014). Learning Phrase Representations Using rnn Encoder-decoder for Statistical Machine Translation, arXiv:1406.1078
  6. Eo KS and Lee KC (2017). Predicting stock price direction by using data mining methods: Emphasis on comparing single classifiers and ensemble classifiers, Journal of The Korea Society of Computer and Information, 22, 111-116.
  7. Hochreiter S and Schmidhuber J (1997). LSTM can solve hard long time lag problems, Advances in Neural Information Processing Systems, 473-479.
  8. Kim E (2021). An integrated model for predicting changes in cryptocurrency return based on news sentiment analysis and deep learning, Information Systems Review, 22, 19-32.
  9. Kim P (2019). An analytical study on automatic classification of domestic journal articles using random forest. JKOSIM, 36, 57-77. https://doi.org/10.5916/jkosme.2012.36.1.57
  10. Lahmiri S and Bekiros S (2019). Cryptocurrency forecasting with deep learning chaotic neural networks, Chaos, Solitons & Fractals, 118, 35-40. https://doi.org/10.1016/j.chaos.2018.11.014
  11. Lee YM, Kwon P, Yu KY, and Kim JY (2017). Method for spatial sentiment lexicon construction using korean place reviews, Journal of the Korean Society for Geospatial Information Science, 25, 3-12.
  12. Park S, Um C, On B, and Jeong D (2019). An AR-KNU sentiment lexicon-based university reputation assessment using online news data, The Journal of Korean Institute of Information Technology, 17, 11-21. https://doi.org/10.14801/jkiit.2019.17.3.11
  13. Rapach D and Zhou G (2013). Forecasting stock returns, Handbook of economic forecasting, 2, 328-383.
  14. Ryu S, Hyun J, and Lee S (2018). An analysis of relationship between social sentiments and cryptocurrency price: An econometric analysis with big data, Information Systems Review, 21, 327-336.
  15. Shapiro AH, Sudhof M, and Wilson DJ (2020). Measuring news sentiment, Journal of Econometrics, In press.
  16. Xu Y (2020). Bitcoin Price Forecast Using LSTM and GRU Recurrent networks, and Hidden Markov Model, University of California, Los Angeles.
  17. Yang M, Xie J, Mao P, Wang C, and Ye Z (2018). Application of the arimax model on forecasting freeway traffic flow, In 17th COTA International Conference of Transportation Professionals: Transportation Reform and Change-Equity, Inclusiveness, Sharing, and Innovation, 593-602, American Society of Civil Engineers Reston, VA.
  18. Zhang W, Wang P, Li X, and Shen D (2018). Some stylized facts of the cryptocurrency market, Applied Economics, 50, 5950-5965. https://doi.org/10.1080/00036846.2018.1488076