Assuming a 1-day VaR holding period, the third stage involvesbootstrapping from our data set of standardised returns: we take alarge number of drawings from this data set, which we now treatas a sample, replacing each one after it has been drawn and mul-tiplying each such random drawing by the volatility forecast 1 dayahead:
rt =
t +
z∗t+1 (11)
where z* is the simulated standardised return. If we take M draw-ings, we therefore obtain a sample of M simulated returns. Withthis approach, the VaR(˛) is the ˛% quantile of the simulated returnsample.