Stock Prediction Model using Seq2Seq and Bi-directional LSTM

Abstract

The stock market is an exchange platform that provides a venue for investors, traders and institutions to buy and sell shares of listed companies. It acts as a surety between the buyer and lender for the smooth exchange of stocks on the accepted price. The stock market sets those accepted price using the mechanism of price discovery based on both fundamental and technical variables The entire idea of predicting stock prices is to gain significant profits. In this paper, a model has been put forward for the prediction of the Closing prices of the stock by using BLSTM based Seq2Seq Model. Seq2Seq model is used as it helps in the mapping of the input-sequence to the output-sequence. This proposed model was compared with various existing algorithms such as K-Nearest Neighbour, Decision tree and Linear Regression. The results shows that the Root Mean Square Error value for the proposed model was found to be least compared to other models.