Quote from sle:
I would take any predictions based on 4 periods with a large brick of salt.
If you look at the long-term history history for Dow Jones IA, median realized volatility (1m rolling frame) is approximately 14%. There have been, however, lengthy periods where realized came in half of the median and where it came in multiple times higher. The in the post-WWII period, for example, the market has realized under 10 vols for over 20 years. Similarly, the market realized significantly higher volatility in the pre-war period, with little respite for over 20 years.
In any case, the experience shows that trying to predict the speed of mean reversion for volatility is super-hard (unless you are an delusional academic). Trying to build some sort of model for the "periodic volatility cycle" is simply impossible.