Share price prediction of Indian Stock Markets using timeseries data-A Deep Learning Approach

Stock markets form the largest avenues of investment in India primarily through two stock exchanges: Bombay Stock Exchange(BSE) and National Stock Exchange(NSE). Analysts and investors look into various factors and try to predict the trends in stock prices in these exchanges. Being extremely volatile in nature, share price prediction is a fairly complex task. Despite the abundance of data, technology has not been able to carry out right predictions up to a desired accuracy most of the time. The recent developments in deep learning technology have proven to be a useful resource in improving the accuracy of predictions. The proposed deep learning based prediction algorithms makes use of Recurrent Neural Network, Long Short Term Memory and Gated Recurrent Unit over time series data obtained online. The developed algorithms predicts the trends five days in advance. The results of the prediction on stocks from various industries are explored to derive valuable insights.