How RiskMetrics 2006 Volatility Differs from Exponentially Weighted Volatility

Summary

The document compares the simple exponentially weighted moving average used in legacy RiskMetrics volatility estimates with the more involved RiskMetrics 2006 approach. The response characterizes the later methodology as combining an ARCH-like process with long memory, a Student distribution, residual scale correction, and lagged correlations between returns. Another answer adds that volatility is computed using scaled one-year daily returns. These changes are presented as part of a framework intended to improve accuracy and support longer risk horizons.

The excerpt does not provide equations, calibration details, or comparative performance evidence, and it does not establish that the older exponential weighting can be substituted without changing results. It points to an introductory treatment elsewhere but does not summarize that treatment. Practitioners considering the method for variance-covariance value-at-risk should consult the full methodology and assess its assumptions, horizon behavior, and data requirements against their own portfolio and validation needs.

Key ideas

• Legacy RiskMetrics volatility uses exponentially weighted returns with a decay factor.
• The 2006 approach is described as combining long memory, a Student distribution, scale correction, and lagged correlations.
• The later method is presented as more complex and suited to longer risk horizons.
• The excerpt supplies no equations or evidence that the legacy estimator is a direct substitute.

Tags

# Difference in Volatility Calculation from RiskMetrics 1996 to RiskMetrics 2006 VaR


# Difference in Volatility Calculation from RiskMetrics 1996 to RiskMetrics 2006 VaR












In the original legacy RiskMetrics documentation from 1996, volatility is calculated using a simple exponentially weighted moving average with some decay factor to determine the weights. This would be used in the calculation of volatility for variance-covariance VaR for example. In RM2006, is volatility still calculated this way? If not, how does it differ and can the 1996 method be substituted?

## Answer by Nicholas (score 2, accepted)

https://quant.stackexchange.com/a/24712

The RiskMetrics 2006 methodology:

> Therefore, the new RM2006 methodology can be summarized as based on a ARCH-like process with long memory + Student distribution + residuals scale correction + lagged correlations between returns. All the ingredients contribute to the performances, albeit possibly at different risk horizons or according to different performance measures. At the end, the new RM2006 is clearly a more complicated methodology than the simple exponential moving average of RM1994. This is the price to pay for the increased accuracies and the possibility to reach long risk horizons

also: A gentle introduction to the RM 2006 methodology.

## Answer by Bob (score 1)

https://quant.stackexchange.com/a/24669

No. Vol in the 2006 is computed using scaled 1 year daily returns.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.