Predicting Stock Prices in the Banking Sector of Bangladesh: A Machine Learning Approach

Abstract Stock price prediction is a daunting task to predict stock prices, particularly in illiquid and highly volatile markets such as the Bangladesh Stock Market, which is dominated by powerful economic, political and market forces that have a significant influence on results. In this paper, sophisticated machine learning models employed to predict stock price are contrasted with the history of previous markets to forecast short-term and long-term direction of prices. Techniques such as Long Short Term Memory (LSTM) networks, Support Vector Regression (SVR), Prophet, and Autoregressive Integrated Moving Average (ARIMA) were employed. Dataset of historical price and trading volumes were utilized to improve prediction accuracy, Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE). The findings indicate that the ARIMA model outperformed the other models being considered. The current paper provides a novel contribution to the field of Bangladesh stock market forecasting and also provides a method for increasing investment strategy through advanced AI techniques.