Direct Answer
Volatility of volatility measures how much a volatility series itself varies, commonly as the standard deviation of changes in volatility or implied volatility. The VVIX index applies the idea to options on the VIX. Higher values mean volatility is unstable: it jumps around instead of drifting.
Volatility of Volatility: Formula, Meaning, Signals, Examples and How to Use It
How Is Volatility of Volatility Calculated?
Volatility of volatility measures how much a volatility series itself varies, commonly as the standard deviation of changes in volatility or implied volatility. The VVIX index applies the idea to options on the VIX.
Formula: vol of vol = standard deviation of ln(sigma_t / sigma_(t-1)) over a window, optionally multiplied by sqrt(252) to annualize
Inputs:
- A series of volatility readings (realized or implied)
- Window length
- Annualization factor if used
Worked example. Four daily log changes in volatility of 5%, -3%, 8% and -6% have a standard deviation of 6.6%. The inputs are illustrative numbers, not market data.
Reading it. Higher values mean volatility is unstable: it jumps around instead of drifting. This matters for options with vega exposure, because uncertain volatility raises the value of convexity such as vomma.
Limits. It needs a clean volatility series to start with, and realized-volatility series are themselves noisy estimates, which inflates the measure. Window length and whether you use log or simple changes matter.
What is Volatility of Volatility?
Volatility of volatility measures how much a volatility series itself varies, commonly as the standard deviation of changes in volatility or implied volatility. The VVIX index applies the idea to options on the VIX.
How to use Volatility of Volatility
Higher values mean volatility is unstable: it jumps around instead of drifting. This matters for options with vega exposure, because uncertain volatility raises the value of convexity such as vomma. A reading alone is never a buy or sell signal; it needs context from price structure, volume, volatility and the broader market regime.
Strengths
- Taking the standard deviation of log changes in volatility puts the instability of volatility in proportional terms.
- It applies to any volatility series, realized or implied, so the same formula covers a range of inputs.
- Reproducible when the formula and data convention are fixed.
Weaknesses and limitations
- Log changes require strictly positive volatility readings, so a zero or missing value breaks the calculation.
- A small base volatility makes log changes large, so low-volatility periods can show high volatility of volatility mechanically.
- Correlated indicators can create false confidence by repeating the same underlying information.
Editorial and risk note
This page is educational content, not individualized investment advice. It should not imply guaranteed prediction, accuracy or outperformance.
Frequently Asked Questions
Is Volatility of Volatility a buy or sell signal?
No. Higher values mean volatility is unstable: it jumps around instead of drifting. It needs a clean volatility series to start with, and realized-volatility series are themselves noisy estimates, which inflates the measure. A trade built on it still needs a trigger, an invalidation rule, a position size and realistic execution assumptions.
What is the best setting for Volatility of Volatility?
Volatility of Volatility takes these inputs: a series of volatility readings (realized or implied); window length; annualization factor if used. There is no universal best setting for the tunable ones, so test them across instruments, regimes and out-of-sample periods and favor stable parameter regions over one historical winner.
Can Volatility of Volatility be used by itself?
It needs a clean volatility series to start with, and realized-volatility series are themselves noisy estimates, which inflates the measure. Window length and whether you use log or simple changes matter. For that reason it is read next to measures of direction, participation, valuation, liquidity and risk, not alone.
Does Volatility of Volatility work on every timeframe?
The calculation (vol of vol = standard deviation of ln(sigma_t / sigma_(t-1)) over a window, optionally multiplied by sqrt(252) to annualize) can be run on any bar size, but behavior is not portable: a reading on a five-minute chart describes a very different market window from the same reading on a daily chart. Validate the exact timeframe before relying on it.
Why does Volatility of Volatility give false signals?
Log changes require strictly positive volatility readings, so a zero or missing value breaks the calculation. A small base volatility makes log changes large, so low-volatility periods can show high volatility of volatility mechanically. Lag, noise and regime changes add to this, because market participants react to new information after the reading is calculated.
How should Volatility of Volatility be backtested?
Rebuild Volatility of Volatility from a series of volatility readings (realized or implied) using point-in-time data, write the rules before testing, include delisted securities where relevant, model realistic fills and costs, reserve a validation sample and compare with a simpler baseline.