Value-at-Risk (VaR) Modeling of LQ45 Stocks Using the GARCH Approach

Saskia Iqlima Bilhaq (Mathematics Undergraduate Study Program, Faculty of Mathematics and Natural Sciences, Padjadjaran University, Sumedang, Indonesia)
Rona Afiya Muthmainah (Mathematics Undergraduate Study Program, Faculty of Mathematics and Natural Sciences, Padjadjaran University, Sumedang, Indonesia)
Moch Panji Agung Saputra (Department of Mathematics, Faculty of Mathematics and Natural Sciences, Padjadjaran University, Sumedang, Indonesia)

International Journal of Quantitative Research and Modeling
Vol. 6 No. 4 (2025): International Journal of Quantitative Research and Modeling

This study aims to model volatility and measure market risk of leading Indonesian stocks included in the LQ45 index using the GARCH approach. Daily closing price data from September 2022 to August 2025 were analyzed to estimate conditional volatility and Value-at-Risk (VaR) at 95% and 99% confidence levels. The GARCH model was selected to capture volatility clustering and conditional heteroskedasticity in stock returns. Residual distributions considered include normal, Student-t, and skewed-t to improve risk estimation, particularly for extreme events. Results indicate that most stocks are best modeled by GARCH(1,1) with a Student-t distribution, reflecting fat tails in return data. VaR estimates provide realistic maximum potential losses varying across stocks, with UNVR and ADRO showing relatively higher risk levels. Backtesting through Kupiec and Christoffersen tests confirms the accuracy and reliability of the GARCH-based VaR model for risk management. This study offers practical insights for investors and portfolio managers in understanding and managing risk exposures of top Indonesian stocks.