Random Thoughts RE: Trading Volatility ETFs (Part 1)

01Mar11

This is the first in a series of random thoughts re: volatility ETF trading.

Readers familiar with modeling these types of strategies won’t find anything earth-shattering here, but for those new to the subject I hope this can be a jumping off point. Note that this is just one humble blogger’s $0.02 – do your own homework and draw your own conclusions.

Thought #1: the VIX is Mean-Reverting

Okay, let’s start with the most obvious …

The VIX itself is strongly mean-reverting, meaning low values tend to portend high values, and vice-versa. This is particularly true when “buying” weakness, because the VIX often spikes up, but doesn’t spike down, making for a favorable risk/reward.

To illustrate, the graph below assumes we could buy (short) the VIX index at the close when the 10-day EMA of the VIX closed under (over) the 10-day SMA. An EMA is a faster moving average than an SMA, so this is a mean-reversion strategy.

[logarithmically-scaled, growth of $]

Note the consistency with which VIX mean-reverts long/short (blue), but also how much more effective that reversion is on the long side (green) than the short (red).

Thought #2: ETFs and the VIX Futures Term-Structure

Of course you couldn’t simply take the above results and assume they could be applied to a volatility ETF like VXX.

As discussed previously, the VIX futures term-structure (which I use loosely to mean many things) acts as a strong tailwind in favor of either long or short volatility ETFs.

For example, over the last few years VIX futures have been mostly contangoed, adding a powerful boost to short volatility positions and making the long trade difficult in anything but the shortest of timeframes (read more).

That means that any mean-reversion play has to respect the term-structure and be biased towards either buying weakness or shorting strength, depending on what the term-structure allows.

Thought #3: Long-term Positions

I’ll break from conventional wisdom and say that simply holding a long-term position in the direction that the term-structure is calling for (which is effectively what investors have been doing with short VXX positions over the last couple of years) will underperform moving forward.

I think VXX’s two year march to zero (which is a result of a bull market that has been near-unprecedented in its plodding consistency) has fooled investors into thinking that these ETFs are appropriate as long-term holds.

But as I hope I demonstrated here, these last two years have been the exception not the norm, and a trend-following or term-structure-following approach isn’t going to be as effective moving forward.

Side note: I do think, however, that something like XVIX (which is basically a 50% short VXX position/100% long VXZ position) might be appropriate as a long-term hold (read more from Only VIX).

Stay Tuned for Part 2

I’m at my self-imposed word limit, so stay tuned for part 2 of this series and more random thoughts re: volatility ETF trading.

Happy Trading,
ms

. . . . .

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

Filed under: VIX & Volatility   |  3 Comments