Market Risk Assessment of the Indonesian Composite Stock Price Index (IHSG) Using Monte Carlo Simulation and Backtesting During the Rupiah Exchange Rate Shock

Abstract

This study aims to quantify the market risk of the Indonesian Composite Stock Price Index (IHSG/JKSE) during a period of heightened macroeconomic uncertainty coinciding with a sharp depreciation of the Indonesian Rupiah (IDR). This research used daily closing prices of IHSG from January 2, 2023 to June 5, 2026 (813 trading days) were used. Missing observations caused by market holidays were imputed via the Last Observation Carried Forward (LOCF) method. Log-returns were computed and their distributional properties examined through descriptive statistics, the Jarque-Bera normality test, and the Augmented Dickey-Fuller (ADF) unit root test. Market risk was estimated as Value-at-Risk (VaR) at the 95% confidence level using Monte Carlo simulation with 10,000 iterations, drawing random returns from a normal distribution parameterized by the empirical mean and standard deviation. Model validity was assessed using Kupiec's proportion-of-failures backtesting procedure. The result shows that the daily IHSG log-return exhibits a negative mean (= -0.000249), volatility of = 0.010851, negative skewness (-1.3644), and total kurtosis of 12.0664 (excess kurtosis = 9.066), confirming a heavy-tailed, non-normal distribution (Jarque-Bera p < 0.001). The ADF test confirms stationarity at the level. The Monte Carlo VaR at the 95% confidence level is -1.8065%, equivalent to a maximum potential loss of IDR 17,902,811 for a one-billion-rupiah portfolio. Backtesting produced exactly 41 actual violations against 41 expected violations (failure rate = 5.04%), confirming that the model is statistically valid and robust for capturing downside market risk.