Analysis of ARIMA-Artificial Neural Network Hybrid Model in Forecasting of Stock Market Returns

Main Article Content

Yakubu Musa
Stephen Joshua

Abstract

This study focuses on the modelling of Nigerian stock market all–shares index and evaluations of predictions ability using ARIMA, Artificial Neural Network and a hybrid ARIMA-Artificial Neural Network model. The ARIMA (1,1,1) model and neural network with architecture (6:1:3) turns out to be the most fitted among the considered models, these models were used for forecasting the returns, and their performances have been compared according to some statistical measure of accuracy. A hybrid model has been constructed using ARIMA-Artificial Neural Networks model, the computational results on the data reveal that the hybrid model using Artificial Neural Network, provides better forecasts, and will enhance forecasting over the single ARIMA and Artificial Neural Networks models. The study recommends the use of ARIMA-Artificial neural network for modelling and forecasting stock market returns.

Keywords:
ARIMA, neural network, hybrid model, forecasting.

Article Details

How to Cite
Musa, Y., & Joshua, S. (2020). Analysis of ARIMA-Artificial Neural Network Hybrid Model in Forecasting of Stock Market Returns. Asian Journal of Probability and Statistics, 6(2), 42-53. https://doi.org/10.9734/ajpas/2020/v6i230157
Section
Original Research Article

References

Bashar A, Maysam A. Predicting financial time series data using hybrid model. Intelligent Systems and Applications. 2016;10:19-41.

Norizan M, Maizah HA. Forecasting Malaysia load using hybrid model. Statistika. 2010;10:1-8.

Ping-Feng P, Chih-Shen L. A hybrid ARIMA and support vector machine model in stock price forecasting. International Journal of Management Science. 2005;33:497-505.

Kyungjoo L, Schwan Y, John JJ. Neural network model vs SARIMA model in forecasting Korean stock market index. Issues in Information System. 2007;2:372-378.

Diaz-Robles LA, Juan C. Ortega, Joshua SF. A hybrid ARIMA and artificial neural networks model to forecast particulate matter in urban areas; The case of Temuco, Chile. Journal Atmospheric Environment. 2008;42:8331-8340.

Akinwale A, Arogundade OT, Adekoya AF. Translated Nigeria stock market prices using artificial neural network for effective prediction. Journal of Theoretical and Applied Information Technology. 2009;2:36-43.

Xiping W, Ming M. A hybrid neural network and ARIMA model for energy consumption forecasting. Journal of Computers. 2012;7:1184-1190.

Cagdas H, Erol E, Cem K. Improvement in forecasting accuracy using the hybrid model of ARFIMA and feed forward neural network. American Journal of Intelligent Systems. 2012;2:12-17.

Isenah GM, Olubusoye EO. Forecasting Nigerian stock market returns using ARIMA and artificial neural network model. CBN Journal of Applied Statistics. 2014;5:25-48.

Suleiman S, Gulumbe SU, Asare BK, Abubakar M. Training dynamic neural networks for forecasting Naira/Dollar Exchange return volatility in Nigeria. American Journal of Management Science and Engineering. 2016;1(1):8-14.

Box GE, Jenkins. Time series analysis forecasting and controls. San Francisco. USA; 1974.

Zhang GP. Times series forecasting using a hybrid ARIMA and neural network model. Neurocomputing. 2003;50:159–175.

Yakubu Musa. Modeling monthly average temperature of Sokoto metropolis using short term memory models. International Journal of Academic Research in Business and Social Sciences. 2014;4(7):382-397.