Abstract:
This paper looks at the application of the artificial neural networks (ANN) in predicting stock market prices in Kenya. In particular the paper looks at the application of ANN in predicting future Equity Bank share prices using historical data. We have assumed that only previous prices affect future prices, then fitted ARIMA models to the stock prices data in order to identify the best input lags into the ANN model. The best combination of lags was taken for input lags and led to optimal result in terms of the least mean squared error between the predicted values and the test data. The 3−3−1 network architecture gave the best results in terms of the Mean Squared error. The paper demonstrates that artificial neural networks can effectively model local stock market prices for reliable forecasts.