Application of Artificial Neural Network to Stock Forecasting – Comparison with SES and ARIMA

Abstract

Stock market also known as equity market is a public entity which is a loose network of economic transactions, not a physical facility or discrete entity for the trading of company stock or shares and derivatives at an agreed price. Artificial Neural Network (ANN) is a field of Artificial Intelligence (AI), which is a common method to identify unknown and hidden patterns in data which is suitable for stock market prediction. In this study we applied a time-delayed neural network model for forecasting future price of stock by using Artificial Neural Network (ANN) methodology. We compared ANN with Single Exponential Smoothening (SES) and Autoregressive-Integrated-Moving-Average (ARIMA) models, the ANN forecasting tool proved to be more precise than the SES and ARIMA as it had a smaller Root Mean Squared Error (RMSE) of 0.686 as compared to the RMSE of the SES which was 2.7400 and ARIMA which was 1.6570.