FRM Part I · FRM Exam Part I · Measuring and Monitoring Volatility
Which statement about GARCH(1,1) parameters is correct?
A GARCH(1,1) model is mean-reverting with a finite long-run variance only when alpha plus beta is below one. Alpha governs reaction to the latest shock, beta governs persistence of past variance, and at a sum of one the model becomes IGARCH without mean reversion.
- AThe model is covariance stationary and mean-reverting only if alpha + beta is less than oneCorrect
- BA larger beta makes variance respond faster to yesterday's return shock
- CIf alpha + beta equals one the model has a finite long-run variance equal to omega
- DAlpha measures the persistence of variance and beta measures the weight on the long-run variance
Explanation
Stationarity with a finite long-run variance omega/(1-alpha-beta) requires alpha+beta<1. Larger alpha, not beta, increases reaction to shocks. At alpha+beta=1 (IGARCH) there is no finite long-run variance and 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 risk manager notes that a GARCH(1,1) model fitted to an equity index has alpha + beta = 1.00 (omega = 0). Which statement is correct about…
- A 20-day equally weighted estimate of daily volatility is 1.2%. A 10-day equally weighted estimate of the same series is 2.0%. Zero mean is …
- A risk manager estimates a two-asset portfolio's variance from volatilities and correlation. Weights are 50% each, volatilities are 10% and …
- Over the past year, equity index options with a strike 10% below spot have had implied volatilities of 28%, while at-the-money options have …
- Daily returns have a standard deviation of 1.00% and a first-order autocorrelation of +0.25, with no higher-order autocorrelation. What is t…
- An analyst estimates daily volatility from five observed daily log returns: +1%, -2%, +3%, 0%, -2%. Assuming the true mean is zero, the anal…