Comparative Study of Composite Stock Price Index (JCI) Prediction Using the Autoregressive Integrated Moving Average (ARIMA) and Facebook Prophet Methods
The current development of information and communication technology has brought significant changes in various sectors, including the financial sector. One of the technologies that is getting more attention is Artificial Intelligence (AI), By using machine learning algorithms and AI data analysis techniques can predict stock prices. This study aims to determine which analysis model is better in predicting the Composite Stock Price Index (JCI) in the last 20 years, the analysis models used are the Autoregressive Integrated Moving Average (ARIMA) and Facebook Prophet. In a comparative analysis of predictive models, ARIMA (0,1,1) was proven to show better performance compared to Facebook Prophet. This can be seen from the lower MAE (Mean Absolute Error), MAPE (Mean Absolute Percentage Error), MASE (Mean Absolute Scaled Error), SMAPE (Symmetric Mean Absolute Percentage Error) and MSE (Mean Squared Error) values in the ARIMA model. These values indicate that ARIMA produces more accurate predictions with a smaller error rate than Facebook Prophet.
Keywords: Composite Stock Price Index, Artificial Intelligence (AI), Autoregressive Integrated Moving Average (ARIMA), and Facebook Prophet
