FRM Part I · FRM Exam Part I · Measuring and Monitoring Volatility
In the EWMA model, which statement about the decay factor lambda is correct?
A lower lambda puts more weight on the most recent squared return, because that weight is one minus lambda. The estimate therefore reacts faster to new shocks and has shorter memory. A higher lambda gives smoother, slower-moving estimates.
- AA lower lambda gives more weight to recent squared returns, so the estimate reacts faster to new shocks.Correct
- BA lower lambda gives more weight to older observations, making the estimate smoother.
- CLambda must be below 0.5 so that weights sum to one.
- DLambda determines the long-run variance to which the estimate reverts.
Explanation
The weight on the latest squared return is (1 - lambda), so a lower lambda raises that weight and shortens memory, making the estimate more responsive. Higher lambda gives smoother estimates. The weights sum to one for any lambda between 0 and 1, and EWMA has no mean reversion.
Did you get it right without looking?
One question tells you little. A timed set on Measuring and Monitoring Volatility shows your real accuracy, how long you take and where you lose marks.
More Measuring and Monitoring Volatility questions
- A GARCH(1,1) model is estimated with omega = 0.000004, alpha = 0.08 and beta = 0.90 for daily returns. What is the implied long-run daily vo…
- A risk analyst computes daily returns on an equity index over a 250-day window. The index closes at 2,000 on day 0 and at 2,100 on day 1. Wh…
- Which statement about implied volatility versus historical volatility is correct?
- Which feature is a recognized weakness of equally weighted moving average volatility estimates compared with exponentially weighted estimate…
- Across strike prices, options on an equity index show implied volatilities that are highest for low-strike out-of-the-money puts and decline…
- A risk analyst uses an EWMA model with decay factor lambda = 0.94. The previous day's volatility estimate was 1.00% per day, and today's obs…