Predicting Stock Price of Dhaka Stock Exchange Using Technical Indicators and Machine Learning Techniques

Predicting stock prices has long been a major area of research in finance and data science. Accurate stock price prediction can provide significant advantages in decision-making for investors, traders, and financial institutions. However, it remains a highly challenging task due to the complex, non-linear, and volatile nature of financial markets. This paper looks into the predictive ability of five machine learning algorithms; Linear Regression, Random Forest Regressor, Gradient Boosting Regressor, XGBoost Regressor and Multilayer Perceptron (MLP) Regressor in predicting the stock prices of three companies of Dhaka Stock Exchange (DSE) of separate sectors: Unilever, Beximco Pharmaceuticals and Grameenphone. Historical stock price data was used along with technical indicators such as EMA, MACD, and RSI. The models were evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE). Experimental results reveal that Linear Regression outperforms others in prediction accuracy for all companies. XGBoost Regressor underperformed compare to other models for Unilever and Grameenphone. In case of Beximco Pharma, Multilayer Perceptron (MLP) Regressor gives poor prediction accuracy.