<![CDATA[Capital market development is one of the main indicators of a country's economy. Therefore, the ability to predict the movement of the Composite Stock Price Index (JCI) is very important for investors and capital market players. The purpose of this research is to develop a JCI model using the Autoregressive Integrated Moving Average (ARIMA) algorithm with the support of RapidMiner software. Simulation test results show that the ARIMA model can be a useful tool for investors and market researchers to make more meaningful decisions. The research methodology used is CRISP-DM (Cross-Industry Standard Process for Data Mining) which includes the following steps: First, collecting historical JCI data from the relevant period. Second, analyze the time series to understand the characteristics of JCI. Third, applying the ARIMA model with parameters (p;2, d;1 q;2) with MSE 2461.572, using statistical analysis methods and evaluating model performance. Fourth, run the ARIMA model on Colabolatory software to predict JCI activity. Fifth, evaluate model performance using metrics. The results of this study have successfully created the best ARIMA model implemented with Colabolatory and can provide accurate information. This research produces a prediction model with evaluation metrics of RMSE 1.43 and MAE 1.13.]]>
Copyrights © 2025
