A VIX History Lesson
By Paul Cherney
Note: Paul Cherney will be on vacation Friday, July 19, and Monday, July 22. His column will return on Tuesday, July 23.
The 10-day exponential moving average of the VIX (a market volatility index) finished Thursday's session near 37.90.
The following is a study, Excessively High VIX Readings, originally published on Standard & Poor's MarketScope on July 18, 2002:
The 10 day exponential moving average of the VIX was 37.32 as of Wednesday's close (07/17/02). I wrote a program which sifted through data all the way back to the beginning of my VIX data (1986). I limited the search to times when the 10-day moving average (exponential) was above 36.99. I took only dates when the VIX crossed below its 10-day moving average (when the 10-day exponential moving average was higher than 36.99).
Since 1986, this has happened only 14 times. If you looked at the price action in the S&P 500 in the 66 trade days which followed such a signal, you would find that every single one of these signals saw the S&P 500 higher as of the 66th trade day after the signal (see the performance figures in the second column of the table below, entitled "At 66 Days.") That's the good news. The bad news is that there is still downside risk after the VIX crosses down through its 10-day exponential moving average.
Please remember, we have not seen the VIX cross below its 10-day exponential moving average yet.
I looked at the best and worst closing performances in the 66 trade days after a cross lower for the VIX. There is downside risk, 11 out of the 14 signals (78.6% of the time) were followed by lower closes. The worst close in the wake of a signal was a closing loss of 6.8% in 1998.
Column 1: Date, the signal date would be the date that the VIX closed under its 10 day exponential moving average and that 10 day M/A had to be above 36.99.
Column 2: At 66 Days, the percentage gain or loss as of the close of trading on the 66th trade day after the signal date.
Column 3: Best Close, the best closing gain (percentage basis) in the 66 trade days after the signal date.
Column 4: Best On, the number of trade days after the signal date that the best closing gain occurred.
Column 5: Worst Close, the worst closing loss (percentage basis) in the 66 trade days after the signal date.
Column 6: Worst On, the number of trade days after the signal date that the worst closing loss occurred.
If you just averaged all the "Worst 66" performances, the average loss was 2.54%.
If you averaged all the percent changes "At 66" the average of all the gains is 11.11%.
If the history of the VIX can in anyway be predictive, this is a good news/bad news scenario. A cross below the 10-day exponential moving average of the close would represent the signal date and I would expect to see something to the upside at that point, but the odds are 11 in 14 that at sometime over the next 66 trade days, the close on the day of the signal date will be undercut on a closing basis. That's the bad news.
The good news is that every single prior signal has seen the S&P 500 higher as of the close of the 66th trade day after the signal date.
This is only one measurement of the market, there are many other factors which could influence prices.
Cherney is chief market analyst for Standard & Poor's