Stock Market Prediction Using a Deep Learning Approach

The objective of this research is to develop a Deep Learning model to forecast the stock price market, by using the variant of Long Short-Term Memory (Deep LSTM). This model predicts close price of the stock for the next day, choosing as inputs the following data: open, high, low, adj close and close prices. We have used the historical stock prices data from yahoo's financial website over 10 years, by choosing two index datasets: Dow Jones Industrial (DJI) and S and P 500 index (daily values). We designed the architectural structure of the deep learning model using eleven LSTM layers and choosing ADAM optimization tools. In order to get effective in the forecasting model, we have tested the network with different iterations and epochs. We have estimated the effectiveness of our proposed model by using the following performance indicators: the mean square error (MSE), the root mean square error (RMSE), the mean absolute percentage error (MAPE) and the accuracy of model. The experimental results clearly show that our Deep LSTM model has the highest accuracy values when comparing with other related researches. Hence, we can conclude that our Deep LSTM model is suitable for accurate prediction of the stock market time series.