Advanced Stock Price Prediction: A Comprehensive Study Using LSTM And GRU Based Machine Learning Technique

The purpose of this paper is to develop an advanced hybrid stock price prediction model by integrating the advantages and benefits of Long Short-Term Memory Model (LSTM) and Gated Recurrent Units (GRU) to facilitate the users in stock market, papers aim is to create an integrated one step solution for stock market prediction. LSTM is highly recommended to capture long-term dependent trends in sequential historical price of a stock whereas GRU handles short-term dependencies and effective in smaller datasets. This paper tries to combine these two models and make a hybrid model for better accuracy and reliability. This hybrid model improves the prediction capability and addresses the limitation which occurs in traditional methods.