VIX Day-of-Week Seasonality Disappearing?

24Oct11

One small addendum to my previous post about VIX day-of-week seasonality, or the tendency for the VIX to reach its highest point of the week on Monday, before falling throughout the week and ending at its lowest point on Friday.

As I mentioned briefly, this pattern, which has been extremely consistent over the last 20+ years, is starting to wane. To illustrate…

In the graph above I’ve shown the rolling 3-year average daily change in the VIX (from the previous close) on the first (grey) and last day of the week (red).

Geek note: results trend-adjusted. See end of post for details.

Note how for most of the last 20+ years, the VIX has hovered around a 2% gain on the first day of the week (grey) and somewhere south of 0% on the last (red).

But also note how over time this gap has narrowed considerably. Today, the first day of the week still exhibits a more positive bias, but the difference between the first and last day of the week is much smaller than in the past.

Note: I haven’t shown it above, but Thursday has also become much less “bearish” for the VIX.

Will VIX day-of-week seasonality disappear entirely?

I find that hard to believe because there are fundamental reasons this seasonality exists (in the “cash” index only mind you).

I think the VIX will always follow this DOW pattern to some degree, but as always, the data will be our guide and I’m always open to being proven wrong.

In my next post I’ll look at day of the week seasonality in the VXX ETF to understand whether seasonality in the VIX carries over. More to follow.

Happy Trading,
ms

Geek note: I adjusted the results in the graph above to discount the impact of the broader VIX trend. From each 3-year first/last-day average I subtracted the average return of all days over that same 3-year period. Also note that all averages are geometric, not arithmetic.

. . . . .

To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.