Estimation and backtesting of financial risk measures in the Vietnamese stock market: the debate between value-at-risk and expected shortfall

This study addresses the critical challenges of accurate risk measurement in financial markets, especially after financial crises. This study aims to enhance risk management in the Vietnamese stock market by focusing on estimating and backtesting value-at-risk (VaR) and expected shortfall (ES) for the VN30 Index. The ARMA-GARCH model is used to forecast VaR and ES, and its effectiveness is compared with that of the historical method. Adhering to the Basel Committee guidelines, this study uses the traffic light approach, Kupiec's proportion of failures test, Christoffersen's interval forecast test, and Acerbi and Szekely's ES tests to validate the models. Results demonstrate that the ARMA-GARCH model offers reliable risk estimates, particularly at the 97.5% and 99% confidence levels, thus surpassing traditional methods. Furthermore, this study provides financial institutions and investors with a practical and robust framework for risk management in emerging markets, thereby supporting a more resilient financial system.