<![CDATA[The Indeks Harga Saham Gabungan (IHSG) is widely recognized as a central indicator of the Indonesian capital market and reflects overall market performance, investor sentiment, and macroeconomic conditions. Accurate forecasting of the IHSG is essential for investors, financial institutions, and policymakers; however, financial time series data are often characterized by non-stationarity and volatility clustering, which limit the effectiveness of conventional forecasting models. This study applies a hybrid Autoregressive Integrated Moving Average–Autoregressive Conditional Heteroskedasticity (ARIMA–ARCH) model to forecast the IHSG by simultaneously modeling the conditional mean and time-varying volatility. The ARIMA model is used to capture linear temporal dependence in the mean process, while the ARCH component addresses heteroskedasticity in the residuals by allowing conditional variance to change over time. Daily IHSG closing price data from September 2024 to September 2025 are analyzed using the Box–Jenkins methodology, including stationarity analysis, model selection, parameter estimation, and diagnostic validation. The empirical results indicate that the hybrid ARIMA–ARCH model provides improved forecasting accuracy compared to a standalone ARIMA model, particularly in periods of heightened market volatility. The ARCH component successfully captures volatility clustering and enables the construction of dynamic volatility-based prediction intervals, offering additional risk-related insights beyond point forecasts. These findings demonstrate that the ARIMA–ARCH framework is effective for modeling IHSG dynamics and can support better risk management, portfolio optimization, and decision-making processes in the Indonesian capital market.]]>
Copyrights © 2026
