Stock Prediction Based on Long-Short Period Prediction (LSPP) Using Machine Learning

Stock price prediction has been an area of interest for investors, traders, and researchers for many years. In the present era, it is very easy to invest money in stock market but it is very difficult and complex to gain more profit due to high volatility and lack of knowledge. The Stockbroker and investor are looking for an accurate stock prediction model to predict the closing price of stocks for both short and long-term periods, so that more profit from stock market can be gained. This paper uses a recurrent neural network (RNN) approach to increase the accuracy for predicting closing price of stock. The proposed approach, LSPP (Long-Short Period Prediction) used a RNN with new function to provide more accurate closing price of stock as compare to existing RNN and LSTM. The results shows that the LSPP model provide a super fit and high prediction accuracy as compared to RNN and LSTM models.