Blog

Paratrade Systems Weekly Research News

19 Sep

Volatility Liftoff!

After months of dull sideways movement, we finally got a breakout in volatility (Vix). It came as it usually does by virtue of a decline in S&P futures. There are a number of questions to consider:

  1. Should we wait a day or two?
  2. Is it big enough to justify buying S&P 500 (futures)?
  3. Is it better to buy after a sustained period of monotony?

Question 1: Is it better to wait after such a jump?

We can look and see if any move above the monthly average will do or if it’s better to wait since a breakout might carry on for a day or two and we don’t want to be the first ones to buy the falling knife.

vixb-blog1

We can see here that we now stand well above the average.

If we buy it when it gets above the average, we get a P&L equal to the red line below. If we wait a day we get the blue line outcome. Waiting a day may seem prudent but it cuts your total points earned since 2004 to 557 from 1020. Waiting two days earns 807 points but the path is no smoother.

vixb-blog2

Q2: Does it help to choose breakouts that are well above the average?

We saw in chart one that we are well above the average plus one standard deviation. Sometimes we get small rises in Vix that will look rather inconsequential. Perhaps if we wait for only the bigger rises then we’ll perform better.

vixb-blog5

The + .5 standard deviation break earns about the same as when we just use the average by itself. The +1 Std. Dev. also earns about the same. Pushing it out to 2 standard deviations is much worse partly because there are far fewer cases. Since there’s no advantage we’ll keep the simple average as our default.

Q3 – Is it better to buy a breakout after a sustained period of boredom?

To answer this, we have to define boredom. We shall simply ask how many days in a row the Vix index has been below its average. We shall then create a simple trading system that buys the market if Vix crosses above its monthly average and prior thereto it had been below that average for 5, 10, and 20 days (in a row). It will exit the day after Vix closes below that 20-day average.

vixblog6

You can see that the system works quite well if we use a setting of 10 days of calm. (The 20 day solution is plotted in gray.) Its worst period was 8/08-2/09 which is no surprise since that volatility explosion lasted so long that your losses would have been sizable. The total profits amount to about 620 S&P points for all trades taken between 2002 and today. Over that time, the market rose 1300 points. These are the conditions that prevailed last week just before Vix exploded.

How does this compare to just buying any Vix breakout. In the chart below we see the P&L of buying all break-outs and selling when Vix calms down and trades back below its average.

vixblog7

Clearly waiting for a period of calmness helps you filter out extreme bear market cases. Overall this system made more money than the filtered cases but no one would prefer it.

Add Comment