2.3.2. Filtered Historical SimulationFiltered Historical Simulation was proposed by Barone-Adesiet al. (1999). This method combines the benefits of HistoricalSimulation with the power and flexibility of conditional volatilitymodels.Suppose we use Filtered Historical Simulation to estimate theVaR of a single-asset portfolio over a 1-day holding period. In imple-menting this method, the first step is to fit a conditional volatilitymodel to our portfolio return data. Barone-Adesi et al. (1999) rec-ommended an asymmetric GARCH model. The realised returnsare then standardised by dividing each one by the correspondingvolatility, zt= (εt/t). These standardised returns should be inde-pendent and identically distributed and therefore be suitable forHistorical Simulation. The third step consists of bootstrapping alarge number L of drawings from the above sample set of standard-ised returns.