Price Prediction Using Time-Series Algorithms for Stocks Listed on Bursa Malaysia

Abstract

Investing in stocks markets is risky. It needs a lot of research and time to make the right investment decision for earning lucrative profits. In this study, we attempted to assist the investors in predicting stock prices. We evaluated five time-series algorithms in this study, i.e., Auto-Regression Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), Seasonal Auto-Regression Integrated Moving Average (SARIMA), Holt Winter and Prophet on six stocks sectors of Bursa Malaysia. Our findings show that ARIMA and LSTM give relatively low errors in predicting stock prices, and ARIMA is better than LSTM in predicting the trend of the stock prices. The results also suggest that ARIMA is suitable for short-term predictions of stock prices.