Stock Market Price Prediction Using LSTM RNN

Financial Analysis has become a challenging aspect in today's world of valuable and better investment. This paper introduces the implementation of Recurrent Neural Network (RNN) along with Long Short-Term Memory Cells (LSTM) for Stock Market Prediction used for Portfolio Management considering the Time Series Historical Stock Data of Stocks in the Portfolio. The comparison of the model with the traditional Machine Learning Algorithms-Regression, Support Vector Machine, Random Forest, Feed Forward Neural Network and Backpropagation have been performed. Various metrics and architectures of LSTM RNN model have been considered and are tested and analysed. There is discussion on how the sentiments of the customer would affect the stocks along with the changes in trends.