The Role of Machine Learning Algorithms in Stock Market Prediction: A Performance Evaluation

The stock market is where people buy and sell shares of publicly listed companies. These stock prices fluctuate up and down based on demand and supply. The other factors include weather, company news reports, changes in management, etc. People can earn a great fortune if they can find a way to predict the future prices of the stock market and accordingly make their investments. Our research aims to find the best model to predict the stock prices effectively. The models we tested include simple regression models and a stacked LSTM (Long-Short-Term Memory) model designed for accurate prediction of stock prices. The reason behind choosing a stacked LSTM model is that it enables us to process time-series data, hence enabling our model to understand the trends better. The metrics used include Mean Square Error (MSE), Mean Absolute Error (MAE), and Root Mean Square Error (RME). The final scores enable us to determine the best model that does not overfit our data, hence enabling us to maximize our profit. Furthermore, the model is tested by recording prices at an interval of 24 hours, 1 hour, and 5 minutes to find a model that gives us the best accuracy.