A Hybrid Method based on Machine Learning to Predict the Stock Prices in Bangladesh

The stock market is a fascinating industry to research. There are several varieties in it. Many specialists have studied and researched the many patterns that the stock market experiences. Because of several relevant elements, the nature of the stock market movement has always been confusing for investors. The stock market prediction has been challenging due to the data's nature, which is very noisy and time-varying. With machine learning, this study intends to minimize the risk of trend prediction drastically. Many academics are investigating machine learning and deep learning as potential methods for predicting stock values. The proposed system employs two techniques: Regression and Recurrent Neural Networks. The goal is to use previous data to calculate the opening price of the next day's stock. The proposed hybrid model is applied to this data and assessed using MSE and RMSE as performance measures to complete this challenge. Our experimental results reveal that the suggested hybrid model is quite accurate. We feel that this hybrid model will be valuable to stock investors in determining the best timing to purchase and sell on the Dhaka Stock Exchange.