access icon openaccess Bitcoin price forecasting method based on CNN-LSTM hybrid neural network model

This is an open access article published by the IET under the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/)
Received 14/10/2019, Accepted 19/11/2019, Published 13/01/2020

In this study, aiming at the problem that the price of Bitcoin varies greatly and is difficult to predict, a hybrid neural network model based on convolutional neural network (CNN) and long short-term memory (LSTM) neural network is proposed. The transaction data of Bitcoin itself, as well as external information, such as macroeconomic variables and investor attention, are taken as input. Firstly, CNN is used for feature extraction. Then the feature vectors are input into LSTM for training and forecasting the short-term price of Bitcoin. The result shows that the CNN-LSTM hybrid neural network can effectively improve the accuracy of value prediction and direction prediction compared with the single structure neural network. The finding has important implications for researchers and investors in the digital currencies market.

Inspec keywords: feature extraction; cryptocurrencies; convolutional neural nets; macroeconomics; recurrent neural nets; economic forecasting; pricing; transaction processing

Other keywords: single structure neural network; Bitcoin price forecasting method; CNN-LSTM hybrid neural network model; feature vectors; long short-term memory; convolutional neural network; short-term price; investor attention; digital currencies market; macroeconomic variables; direction prediction; transaction data; value prediction; feature extraction

Subjects: Financial computing; Neural computing techniques; Data handling techniques

References

    1. 1)
      • 8. Da, Z, Engelberg, J, Gao, P.: ‘In search of attention’, J. Finance., 2011, 66, (5), pp. 1461–1499.
    2. 2)
      • 6. Jing, L.: ‘Using BP neural network to construct Bitcoin market forecasting model’, Finance Account. Month., 2016, 536, (21), pp. 33–36.
    3. 3)
      • 1. Eom, C., Kaizoji, T., Kang, S.H., et al: ‘Bitcoin and investor sentiment: statistical characteristics and predictability’, Physica A, Stat. Mech. Appl., 2019, 514, pp. 511–521.
    4. 4)
      • 10. Hochreiter, S., Schmidhuber, J.: ‘Long short-term memory’, Neural Comput., 1997, 9, (8), pp. 1735–1780.
    5. 5)
      • 7. Mallqui, D.C.A., Fernandes, R.A.S.: ‘Predicting the direction, maximum, minimum and closing prices of daily Bitcoin exchange rate using machine learning techniques’, Appl. Soft Comput. J., 2019, 75, pp. 596–606. http://www.sciencedirect.com/science/article/pii/S1568494618306707.
    6. 6)
      • 4. Huang, J.Z., Huang, W, Ni, J.: ‘Predicting Bitcoin returns using High Dimensional technical indicators’, J. Finance Data Sci., 2018, 5, pp. 140–155, http://www.sciencedirect.com/science/article/pii/S2405918818300928.
    7. 7)
      • 5. Qiu, L., Fanbin, W., Longxiu, Y., et al: ‘Prediction of Bitcoin price based on wavelet analysis’, Neijiang sci. Technol., 2015, 36, (5), pp. 129–130.
    8. 8)
      • 2. Vassiliadis, S., Papadopoulos, P., Rangoussi, M., et al: ‘Bitcoin value analysis based on cross-correlations’, J. Internet Bank. Commer., 2017, 22, (S7), p. 1.
    9. 9)
      • 3. Ciaian, P., Rajcaniova, M., Kancs, D.: ‘The economics of BitCoin price formation’, Appl. Econ., 2016, 48, (19), pp. 1799–1815.
    10. 10)
      • 9. Yizhe, L.: ‘The study of investor attention's impact on the stock performance based on Baidu Index’ (Hunan University, People's Republic of China, 2015).
http://iet.metastore.ingenta.com/content/journals/10.1049/joe.2019.1203
Loading

Related content

content/journals/10.1049/joe.2019.1203
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading