Stock Market Price Movement Forecasting on BURSA Malaysia using Machine Learning Approach

Stock Markets are volatile, emphasizinsg the need and interest of accurate stock prediction models within the market. Utilizing machine learning models, stock trends prediction and analysis is an area that shows great promise. In this paper, four models - Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), LSTM-GRU and Auto-Regressive Integrated Moving Average (ARIMA) are built and compared to determine which is more suitable for stock prediction. A dashboard is also built for better visualization for end-users. Performance evaluation of the models are relying on both the results of the evaluation metrics and the projected prediction adjacent with the actual pricing of the observed stock data.