Abstract
In this paper, simulation techniques are used to estimate value-at-risk of
the CARBS equity indices and a global minimum variance portfolio. The empirical
analysis in this paper is divided into two parts, the first part deals with simulating
normally distributed returns in order to estimate VaR. In the second part calibrated
univariate GARCH models are used to simulate returns series that are consistent
with the stylised facts of financial time series. When a normal distribution is assumed,
the GARCH model forecast of the returns produces the most reliable result.
Finally, when garch processes are simulated, the EGARCH model is superior.