Smart Media Journal (스마트미디어저널)

THE KOREAN INSTITUTE OF SMART MEDIA (한국스마트미디어학회)
권호 표지
DOI QR코드

DOI QR Code

Financial Market Prediction and Improving the Performance Based on Large-scale Exogenous Variables and Deep Neural Networks

대규모 외생 변수 및 Deep Neural Network 기반 금융 시장 예측 및 성능 향상

  • 천성길 (인하대학교 컴퓨터공학과 대학원) ;
  • 이주홍 (인하대학교 컴퓨터공학과) ;
  • 최범기 (인하대학교 컴퓨터공학과) ;
  • 송재원 (밸류파인더스)
  • Received : 2020.07.28
  • Accepted : 2020.10.12
  • Published : 2020.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Attempts to predict future stock prices have been studied steadily since the past. However, unlike general time-series data, financial time-series data has various obstacles to making predictions such as non-stationarity, long-term dependence, and non-linearity. In addition, variables of a wide range of data have limitations in the selection by humans, and the model should be able to automatically extract variables well. In this paper, we propose a 'sliding time step normalization' method that can normalize non-stationary data and LSTM autoencoder to compress variables from all variables. and 'moving transfer learning', which divides periods and performs transfer learning. In addition, the experiment shows that the performance is superior when using as many variables as possible through the neural network rather than using only 100 major financial variables and by using 'sliding time step normalization' to normalize the non-stationarity of data in all sections, it is shown to be effective in improving performance. 'moving transfer learning' shows that it is effective in improving the performance in long test intervals by evaluating the performance of the model and performing transfer learning in the test interval for each step.

미래의 주가를 예측하기 위한 시도는 과거부터 꾸준히 연구되어왔다. 그러나 일반적인 시계열 데이터와 달리 금융 시계열 비정상성(non-stationarity)과 장기 의존성(long-term dependency), 비선형성(non-linearity) 등 예측을 하는 것에 있어서 여러 가지 방해 요인이 존재한다. 또한, 광범위한 데이터의 변수는 기존에 사람이 직접 선택하는 것에 한계가 있으며 모델이 변수를 자동으로 잘 추출할 수 있도록 하여야 한다. 본 논문에서는 비정상성 데이터를 정규화할 수 있는 슬라이딩 타임스텝 정규화(sliding time step normalization) 방법과 LSTM 형태의 오토인코더(AutoEncoder)를 사용하여 모든 변수로부터 압축된 변수로 미래 주가를 예측하는 방법, 기간을 나누어 전이 학습을 하는 이동 전이 학습(moving transfer learning)을 제안한다. 또한, 실험을 통하여 100개의 주요 금융 변수들만을 사용하는 것보다 뉴럴 네트워크를 통해서 가능한 많은 변수를 사용하였을 때 성능이 우수함을 보이며, 슬라이딩 타임스텝 정규화 방법을 사용하여 모든 구간에서 데이터의 비정상성에 대해 정규화를 수행함으로써 성능 향상에 효과적임을 보인다. 이동 전이 학습 방법은 스텝 별 테스트 구간에서 모델의 성능을 평가하고 전이학습을 함으로써 긴 테스트 구간에서 성능 향상에 효과적임을 보인다.

Keywords

References

  1. Ayodele A. Adebiyi ,Aderemi O. Adewumi, Charles K. Ayo, "Stock Price Prediction Using the ARIMA Model," 2014 UKSim-AMSS 16th International Conference on Computer Modelling and Simulation, Cambridge, UK, 2014.
  2. Keerti Gururaj Kulkarni, Gururaj Anand Kulkarni, "FUNDAMENTAL ANALYSIS vs. TECHNICAL ANALYSIS: A CHOICE OF SECTORAL ANALYSIS," IJEMS, vol. 4, no. 2, pp. 234-246, 2013.
  3. Xiangzhou Zhang, Yong Hu, Kang Xie, Shouyang Wang, E.W.T.Ngai, Mei Liu, "A causal feature selection algorithm for stock prediction modeling," Neurocomputing, vol. 142, no. 22, pp. 48-59, 2014. https://doi.org/10.1016/j.neucom.2014.01.057
  4. Isaac Kofi Nti, Adebayo Felix Adekoya, Benjamin Asubam Weyori, "Random Forest Based Feature Selection of Macroeconomic Variables for Stock Market Prediction," AJASSP, vol. 16, no. 7, pp. 200-212, 2019.
  5. Nishchal Sharma, Dr. Chaman S. Chauhan, "Stock Market Prediction with Artificial Neural Network and Machine Learning: A Review of Literature," IJIRCCE, vol. 4, no. 6, pp. 10360-10363, 2016.
  6. Yakup Kara, Melek Acar Boyacioglu, "Predicting direction of stock price index movement using artificial neural networks and support vector machines: The sample of the Istanbul Stock Exchange," Expert Systems with Applications, vol. 38, no. 5, pp. 5311-5319, 2011. https://doi.org/10.1016/j.eswa.2010.10.027
  7. AH Moghaddam, "Stock market index prediction using artificial neural network," Journal of Economics, Finance and Administrative Science, vol. 21, no. 41, pp. 89-93, 2016. https://doi.org/10.1016/j.jefas.2016.07.002
  8. Kai Chen, "A LSTM-based method for stock returns prediction: A case study of China stock market," 2015 IEEE International Conference on Big Data (Big Data), Santa Clara, CA, USA, 2015.
  9. David M. Q. Nelson, "Stock market's price movement prediction with LSTM neural networks," 2017 International Joint Conference on Neural Networks (IJCNN), Anchorage, AK, USA, 2017.
  10. Dev Shah, Wesley Campbell, Farhana H. Zulkernine, "A Comparative Study of LSTM and DNN for Stock Market Forecasting," 2018 IEEE International Conference on Big Data (Big Data), Seattle, WA, USA, USA, 2018.
  11. Sheng Chen1, Hongxiang He, "Stock Prediction Using Convolutional Neural Network," IOP Conference Series: Materials Science and Engineering, 435, Shanghai, 2018.
  12. Ehsan Hoseinzade, Haratizadeh Haratizadeh, "CNNpred: CNN-based stock market prediction using a diverse set of variables," Expert Systems with Applications, vol. 129, pp. 273-285, 2019. https://doi.org/10.1016/j.eswa.2019.03.029
  13. Jithin Eapen, Doina Bein, Abhishek Verma, "Novel Deep Learning Model with CNN and Bi-Directional LSTM for Improved Stock Market Index Prediction," 2019 IEEE 9th Annual Computing and Communication Workshop and Conference (CCWC), Las Vegas, NV, USA, USA, 2019.
  14. Taewook Kim, Ha Young Kim, "Forecasting stock prices with a feature fusion LSTM-CNN model using different representations of the same data," PLoS ONE, vol. 14, no. 2, 2019.
  15. Ashish Vaswani, Noam Shazeer, Niki Parmar, "Attention Is All You Need," arXiv:1706.03762, 2017.
  16. Kelvin Xu, Jimmy Ba, Ryan Kiros, Kyunghyun Cho, "Show, Attend and Tell: Neural Image Caption Generation with Visual Attention," 1502.03044, 2015.
  17. Hwang Jun Kim, "Detection and Damping Recognition of Normal Frequency Using Fast Fourier Transform in the Vibration Acceleration Analysis System," Smart Media Journal, vol. 8, no. 2, pp. 16-20, 2019.
  18. Dong Ju Park, Byeong Woo Kim, Young-Seon Jeong, Chang Wook Ahn, "Deep Neural Network Based Prediction of Daily Spectators for Korean Baseball League : Focused on Gwangju-KIA Champions Field," Smart Media Journal, vol. 7, no. 1, pp. 16-23, 2018. https://doi.org/10.30693/SMJ.2018.7.1.16
  19. Yao Qin, Dongjin Song, "A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction," IJCAI, pp. 2627-2633, 2017.
  20. Inbok Lee, "Order preserving matching with k mismatches," Smart Media Journal, vol. 9, no. 2, pp. 33-38, 2020. https://doi.org/10.30693/SMJ.2020.9.2.33