Stats › Volatility and VIX › Implied against realised
The volatility index overstates the next month by four to five points, and nobody has checked since 2017
| Forecast | Median error | Average | 25th percentile | 75th percentile |
|---|---|---|---|---|
| Raw volatility index | 4.62 | 5.27 | 2.65 | 7.01 |
| Recent realised volatility | 3.06 | 4.25 | 1.37 | 5.66 |
| Mean reversion adjustment | 3.02 | 4.08 | 1.47 | 5.20 |
| An adjusted index measure | 2.31 | 3.58 | 1.09 | 4.45 |
Absolute difference between the forecast and the subsequent realised volatility, 21 December 1990 to 30 September 2017. The raw index is the worst of the four forecasts on every statistic shown. The publisher reports no significance test, no regression coefficient, no r squared and no correlation anywhere in this document, and gives no average index level and no average realised volatility level.
Over 1990 to 2017 the volatility index exceeded subsequent thirty day realised volatility by four to five points on average, with a median absolute forecast error of 4.62. No opened study measures the same bias on a sample ending after October 2017.
Members only
The rest of this page is for members
Below this point there are 5 sections and 4 named sources, roughly 800 words of it. Every figure carries the source it came from and the date the data is from.
You can keep browsing every statistic in the library for free. The intro and the summary are always open.