Daily updated volatility and VaR forecasting
Prior to the 2008 crisis linear factor models typically focused on long-term risk estimation while the need for short time horizon VaR numbers was usually met using an exponentially weighted moving average (EWMA) approach based on up to 100 days of historic data.
While both approaches were reporting low levels of risk prior to the crisis, the EWMA models responded more quickly to the changing market environment.
This lead to a greater interest in being able to estimate short term risk – up to 1 month ahead – and was encouraged by regulatory developments such as the EU’s UCITS rules which required managers to calculate a near-term risk estimate and confirm its performance against outcomes.
An additional advantage of the EMA model construction approach is that it is possible to estimate the daily performance of the estimated factors out of sample by deduction from the price performance of a large number of liquid assets. By applying a GARCH process to these daily returns a near-term variance forecast for each factor can be generated. This offers a true forward looking set of factor variances and covariances and is valuable in both risk analysis and efficient portfolio construction. The use of GARCH has important advantages over EWMA as it allows for forecasts at different time horizons – say 1 day or 20 days – that are not simple scaled values but which take account of the tendency of volatility to revert to trend over time.
EMA has gone further and incorporated a residual scaling factor into the forecasts to take account of the tendency for residuals to increase across the board during high volatility periods.
The FASTVaR forecasts provide a credible alternative to reliance on market level implied volatility measures which have themselves become tradeable and subject to their own specific events.