A Hybrid Approach of Combining BP Neural Network and GARCH Model for Forecasting Stock Price

DOI

10.2991/cnci-19.2019.32How to use a DOI?

Keywords

BP neural network, Stock price prediction, Nonlinear, ARIMA, GARCH.

Abstract

In this paper, we first construct a three-layer (one hidden layer) multilayer back propagation neural network (BPNN) model to forecast daily closing prices of stocks, but there are considerable errors between the actual values and predicted values. Then, to get better prediction results with higher accuracy, we fit the tendency of the errors by modeling a generalized autoregressive conditional heteroscedasticity (GARCH) model. Since it can better deal with the non-linearity and other characteristics of financial data, so the predictive effect of our method is better than that of the hybrid approach of BPNN and autoregressive integrated moving average (ARIMA) model. Finally, we verify this assertion through experimental results.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Peipei Zhang
AU  - Chuanhe Shen
PY  - 2019/05
DA  - 2019/05
TI  - A Hybrid Approach of Combining BP Neural Network and GARCH Model  for Forecasting Stock Price
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 222
EP  - 226
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.32
DO  - 10.2991/cnci-19.2019.32
ID  - Zhang2019/05
ER  -