Chandrika, P.V., Sakthi Srinivasan, K.
Stock markets are found to be non-volatile in nature. To understand the volatility in the markets one requires certain techniques to be adopted. With the emergence of Artificial Intelligence various techniques have been adopted to study the volatility in stock markets. Many researcher applied techniques of machine learning and artificial intelligence to predict the stock market prices. Knowing the direction of the stock market helps the investors to plan their investments and boost their returns. There is an enormous study done in the area of predicting the stock market prices of the companies by adopting these techniques. But the present study considers the market indices from developed and emerging markets which constitute NASDAQ Composite Index from United States, KOSPI Market Index from South Korea and tries to apply the deep learning neural network model of Recurrent Neural Network with Long Short Term Memory (RNN+LSTM). The selected indices are modelled on RNN+LSTM to predict the direction of the market index. Data considered for the study ranges from Jan 2012 to March 2021 capturing the variables Low Price, High Price, Open Price, Close Price and Volume on each day. One day before closing price of the index is fed as input and is processed onto Recurrent Neural Network with Long Short Term Memory to predict one day a head direction of the index. The performance of the LSTM model is evaluated based on the accuracy of prediction which is found to be 81% on NASDAQ and 81.9% on KOSPI market Index.