An empirical comparison of GARCH models based on intraday Value at Risk

In the last decade, one of the main issues of financial study has been risk management as typified by Value at Risk (VaR). Meanwhile it grows popular that financial practicians, from institutional investors to day traders, employ intraday data for making investment decisions. In the light of these developments, it is important to analyze an intraday downside risk. In this report we execute an empirical comparison of univariate/multivariate GARCH models based on intraday VaR. After some necessary data preprocessing such as the extraction of equal interval observation and the adjustment of intraday seasonality, we examine the forecasting performance of one-step-ahead VaR in the following way. First we estimate the parameters of models for a certain period. Second, a VaR process is simulated through the estimated parameters, and the failure rate of risk is calculated that is how frequently the actual return exceeds the VaR percentile. Finally, a likelihood ratio test on the exceedance rate is performed based on the simulation outcome.