Recurrent Neural Network Estimator for Stock Price

This paper provides a solution to one of the biggest challenges of the profit investors which is the stock market prediction. It focuses on forecasting the stock market price based on the principle of Convolutional Neural Network (CNN) architecture that helps in feature extraction of data, adopting Long Short Term Memory (LSTM) for predicting the stock market and using a Kalman Filter to obtain a high accuracy prediction model. The model is evaluated on the stock market data of Apple Inc. and S&P 500. The result has shown that it is reliable to use a combination of CNN-LSTM and Kalman filter to predict the stock price with high prediction accuracy.