Analisis Volatility Clustering dan Karakteristik Tail Risk pada Return Emas Antam Indonesia

Abstract

This study aims to analyze volatility clustering and tail risk characteristics in Indonesian Antam gold returns. The data used in this study consist of daily Antam gold prices, which are transformed into logarithmic returns. The analytical methods include descriptive statistics, the Jarque-Bera normality test, the Augmented Dickey-Fuller (ADF) stationarity test, Autocorrelation Function (ACF) analysis, the Ljung-Box test, the ARCH-LM test, and tail risk measurement using Historical Value-at-Risk (VaR) and Expected Shortfall (ES). The results show that Antam gold returns are not normally distributed, exhibit fat tails, and indicate the presence of volatility clustering and ARCH effects. The ADF test confirms that the return series is stationary. The VaR and ES measurements indicate potential extreme losses at the 1% and 5% risk levels, while the exceedance results, which are close to the expected risk levels, suggest that the historical VaR approach is reasonably relevant in describing tail risk. These findings confirm that although Antam gold is often considered a safe-haven asset, volatility risk and extreme losses should still be considered in investment decision-making. In addition to providing a statistical overview of Antam gold risk, this study also contributes to the field of Data Science in the context of Computational Finance by providing an analytical foundation for risk feature construction, volatility signal identification, and the development of risk prediction models based on artificial intelligence or machine learning.