Macro Investor’s VIX Trading Strategy

By reader request, this is a test of a strategy from Macro Investor for trading VIX ETPs like XIV and VXX. This strategy, like many you’ll find on this blog and elsewhere, uses the shape of the VIX futures term-structure, going short the VIX when futures are more contangoed and long when more backwardated.

Strategy results from 07/2004 trading XIV (inverse VIX) and VXX (long VIX) follow. Read about test assumptions, or get help following this strategy.

20141209.01

The strategy rules I’ve tested are a bit different than the ones presented by Macro Investor in order to stay more in line with how we generally test things here at VMS. Strategy rules:

  • At the close, calculate R, where R = [average of (VX1 / VIX, VX2 / VX1) – 1]. VX1 and VX 2 represent first and second month VIX futures, and VIX the VIX spot.
  • At the close, calculate the average value for R from inception up to that moment in time.
  • Go long XIV at the close if R > (average historical R * -1), otherwise go long VXX. Hold until a change in position.

Differences between my test and Macro Investor’s:

  • MI’s test traded UVXY (2x leveraged VIX) in place of VXX. This strategy has only spent about 6% of all days long VIX. That greatly increases the risk of overfitting those particular days. We would compound that impact of overfitting by assuming that we traded a leveraged product. Bear in mind that VIX ETPs are already insanely volatile. It would be bad stewardship on my part to even share such a backtest.
  • After applying the strategy to VIX ETPs, MI went on to then apply the strategy to ETP options. I have not done so here as options are beyond the scope of what we usually discuss at VMS, but note that my comments below would apply to options trading as well.

20141209.02

20141209.03

I think the basic concept behind the strategy, going short the VIX when futures are closer to contangoed and long when closer to backwardation, is a sound one and we’ve tested a number of similar strategies previously, such as the VIX vs Front Month and First vs Second Month strategies.

On paper, Macro Investor’s strategy has performed better than most, with an exceptionally smooth equity curve after the 2007/08 crash, but I think there’s an extra helping of overfitting with this strategy as a result of the introduction of the “average historical R”.

The average R value is based on all data from inception up to date. Note that the strategy doesn’t “peek” into the future as the average R value is recalculated each day in our test using only the data available to us at that moment in time.

But let’s pretend that we did have perfect foresight and could use the most up-to-date average R value (4.8%) for our entire test. In other words, from day 1 until the most recent trading day, average R always equals ~4.8%.

With this perfect foresight, this strategy should to some degree improve if the average R is actually honing in on some optimal value, but in fact it does just the opposite. It goes from a sexy Sharpe/UPI of 1.3/2.7 to a less impressive 1.0/1.9.

That tells me that the way in which the “average R” has changed over time (which is an entirely arbitrary result of the date on which the test started, and shouldn’t in and of itself be predictive), is overfit to this particular moment in history and unlikely to provide value out-of-sample.

Again though, beyond my criticism re: the average R and use of a leveraged long VIX ETP, I think the basic concept behind the strategy does have value.

* * *

A big thank you to Macro Investor for posting this strategy.

When the strategies that we cover on our blog (including this one) signal new trades, we include an alert on the daily report sent to subscribers. This is completely unrelated to our own strategy’s signal; it just serves to add a little color to the daily report and allows subscribers to see what other quantitative strategies are saying about the market.

Click to see Volatility Made Simple’s own elegant solution to the VIX ETP puzzle.

Good Trading,
Volatility Made Simple

VIX Trading Strategies in November

We’ve tested 20 simple strategies for trading VIX ETPs on this blog (separate and unrelated to our own strategy). And while I can’t speak for all traders, based on all of my readings both academic and in the blogosphere, the strategies we’ve tested are broadly representative of how the vast majority of traders are timing these products.

Most of these strategies turned in a strong performance for the month, making up for broad losses in October. Below I’ve shown the November and YTD results of the 20 strategies we’ve blogged about previously, trading XIV and VXX. Read about test assumptions or get help following these strategies.

20141204.01

20141204.02

Stategies included in this report: First vs Second Month Futures, VIX vs Front Month Futures, VIX vs 1-Month Constant Maturity, VIX vs VXV Indices, V&M’s VIX:VXV Ratio, TM’s RSI(2)DFTB’s StDev Strategy, DFTB’s Spread Strategy, DDN’s Momentum Rotation, DDN’s VRP Strategy, 10/100-Day MA Crossovers, TWP’s Quadratic Fit, NAS’s VIX Futures Momentum, S&P 500 50/200-Day Crossovers, Brute Force Optimized VRP, LI’s Bollinger Band Strategy, LI’s SMA Crossovers, Evolution Capital Strategy, TTO’s VRP Strategy, MS’s Mean-Reversion

MarketSci’s Mean-Reversion was the top performer this month by a slim margin. The real standouts this year though have been the “VRP” strategies tested here and here with YTD returns of 66% and 108% respectively.

I would strongly caution however that when tracking as many different strategies as we track, trading vehicles as volatile as VIX ETPs, something is bound to appear to have the hot hand at any given moment. Whether or not this is a result of real predictive ability or just the luck of the draw is open to interpretation, but what I do know (and I’ve shown quantitatively here and here) is that chasing the most recent top performer is generally a bad idea.

The strategies that are leading the pack today very often find themselves at the back of the pack tomorrow. These VRP strategies are a good example of that. Despite their strong performance this year, both trailed buy & hold badly for all of 2012 and 2013. That’s why I always advocate for a more holistic approach that considers many of the different concepts we’ve discussed on this blog rather than marrying any one single idea.

XIV and VXX are of course not the only show in town. Below I’ve rerun the same tests, this time applying each strategy to the less popular (or is it “underutilized”?) mid-term VIX ETPs ZIV and VXZ (click to zoom).

20141204.03     20141204.04

Note that when any of these strategies signal new trades, we include an alert on the daily report sent to subscribers. This is completely unrelated to our own strategy; it just serves to add a little color to our daily report and allows subscribers to see what other quantitative strategies are saying about the market.

Good Trading,
Volatility Made Simple

MarketSci’s Mean-Reversion VIX Trading Strategy

This is a test of a strategy from MarketSci that uses moving average crossovers to trade VIX ETPs like XIV and VXX. Unlike most of the strategies that we cover on our blog, this strategy is of the mean-reversion variety. Like the RSI(2) strategy that we’ve covered previously, it’s based on going long vol when the VIX is oversold, and short when the VIX is overbought.

Strategy results trading XIV and VXX from 07/2004 follow. Read about test assumptions, or get help following this strategy.

20141121.01

The strategy is meant as a filter to an existing longer-term strategy (again, like the RSI(2) strategy we’ve covered previously). That’s because, while the VIX index itself is predictably mean-reverting, VIX futures and ETPs (a) trade out much of this mean-reversion, and (b) introduce other unique factors like volatility risk premium that sometimes dwarf this mean-reversion effect.

Here we apply MarketSci’s filter to our own strategy. Strategy rules:

  • When our strategy is signaling a short vol position (ex. XIV) for today’s close, and the 10-day EMA of the VIX will close above the 10-day SMA at today’s close, we go long XIV at today’s close.
  • When our strategy is signaling a long vol position (ex. VXX) for today’s close, and the 10-day EMA of the VIX will close below the 10-day SMA at today’s close, we go long VXX at today’s close.
  • Hold until a change in position. If none of the above conditions are met, move to cash. Read about test assumptions, or get help following this strategy.

Because a 10-day EMA is “faster” than a 10-day SMA, this is a mean-reversion strategy, only taking short vol positions when the VIX is overbought, and vice-versa.

Just as important as the filter’s performance, is the performance of all of those trades it would have skipped. In the graph below I’ve compared the trades the filter would have agreed with (and allowed) in blue, versus those it would have disagreed with (and skipped) in grey.

20141121.02

Despite spending a similar amount of time invested, the strategy would have been more productive when MarketSci’s filter agreed with the trade (to the tune of about 58% vs 22% annualized), but the filter would have still left a considerable amount of return on the table. But that’s only part of the story. Below I’ve shown statistics per trade.

20141121.03

Note the very different results for % winning trades and W/L ratio. When MarketSci’s mean-reversion filter agreed with the trade, it was much more likely to be a winner (see win %), but when it was wrong, it tended to be very wrong (see W/L ratio).

That makes sense. Mean-reversion tends to act like a rubber band. When stretched, the rubber band bounces back over and over consistently. But when stretched a little too far, the rubber band (and your portfolio) breaks.

So is there value in MarketSci’s filter?

At least as I’ve applied it here, I wouldn’t trade it as a standalone strategy, as there’s just too much productive time spent out of the market, but I do think there’s value in capturing whether we’re in an overbought or oversold environment because, as we’ve shown here, it changes expectations for both how often a trade might succeed and how badly it might move against us when it fails.

* * *

A big thank you to MarketSci for posting this strategy.

When the strategies that we cover on our blog (including this one) signal new trades, we include an alert on the daily report sent to subscribers. This is completely unrelated to our own strategy’s signal; it just serves to add a little color to the daily report and allows subscribers to see what other quantitative strategies are saying about the market.

Click to see Volatility Made Simple’s own elegant solution to the VIX ETP puzzle.

Good Trading,
Volatility Made Simple

VIX Trading Year-to-Date in Four Graphs

VRP Explained 2Simply looking at a graph of a VIX ETP like VXX or XIV doesn’t tell us much about the underlying forces at play in the VIX complex. ETP prices are merely the end result of the relationship between these underlying forces (which is why traders shouldn’t rely too heavily on price to guide trades, the way one might with say a stock index).

In the next four graphs, I show four of these key relationships year-to-date. These relationships are numbered 1-4 in the image to the right, which shows how volatility, the VIX, and VIX futures are most commonly aligned.

Long-time readers will note that these are the same four I covered in Four Graphs to Rule Them All as far back as 1986. Here I’m updating these graphs YTD to show our recovery from the major VIX spike in October.

20141113.01

Historical volatility (shown here as 10-day annualized standard deviation) is approaching baseline levels of the year, but still a ways off from late-August/early-September when HV approached 20+ year lows.

20141113.02

20141113.03

VIX futures have behaved mostly as you’d expect this year relative to the spot VIX. There has been a fairly constant premium over the spot during periods of market calm (i.e. contango), and futures have trailed the spot during significant spikes (i.e. backwardation, betting on mean-reversion to pull the spot VIX down).

20141113.04

As I noted in Four Graphs to Rule Them All, as you move from graph #1 to graph #4, the relationships illustrated become more and more important, but less and less consistent and/or predictable.

The fourth graph, VIX futures vs the future realized VIX, is really the key to the VIX trading game. Short-term VIX ETPs like XIV and VXX (for example) are perpetually shifting towards the second month contract to maintain a 30-day constant maturity. As long as futures are consistently overestimating the subsequent realized VIX, there will be money to be had in this trade as VIX futures are forced to converge to the VIX spot as they approach expiration.

VIX ETPs have been flat this year because, with the exception of April/May, the VIX complex has failed to maintain a consistent premium (positive or negative) between futures and the subsequent realized VIX. As that is a forward-looking relationship (futures today versus the spot in the future), I chalk that failure up to “stuff happens” rather than any fundamental shift in the VIX complex.

Where do we stand now? Despite VIX futures declining significantly since the major spike in October, futures are still higher than about half of VIX spot readings since the current bull run began in late-2011, and higher than 76% of spot readings this year. That means that, if this market can continue to remain reasonably calm, there’s plenty of room for futures to fall, and XIV/ZIV to rise.

With October’s VIX spike fresh in investors’ minds though, it’s yet to be seen whether we can return to the ultra-low VIX (< 12) seen this summer. That isn’t necessary for inverse VIX ETPs to gain mind you as premium will still exist with the VIX spot staying right where it’s at today, but it will make things more difficult.

Click to see Volatility Made Simple’s own elegant solution to the VIX ETP puzzle.

Good Trading,
Volatility Made Simple

The First Shall Be Last and The Last Shall Be First

A common theme in my writing recently has been not being married to any one specific set of trading rules. It’s very easy to get excited about an amazing recent performance (such as the Brute Force VRP strategy in October), but over the long-term, I’m sure that a holistic approach that considers many different concepts will win the day.

This year has been a perfect illustration of that. Every month I produce a graph like the one below showing the year-to-date performance of all of the simple strategies that we’ve blogged about.

20141106.01

Read about test assumptions, or get help following these strategies.

That’s a lot of information to digest, so in the next graph I’ve stripped out just the best performing strategy YTD in blue (Brute Force VRP), and the worst performing in grey (DFTB’s StDev).

At near-term peaks at points #1 and #2, DFTB’s StDev was the best performing strategy of all of those we track, while the Brute Force VRP strategy was either the absolute worst, or somewhere very close.

Fast-forward to point #3 where we find ourselves today, and all of that has been flipped on its head. DFTB’s StDev is now at the bottom of the barrel, and the Brute Force VRP strategy is the tippity-top performer.

20141106.02

Here’s the rub though, there really is no way to know in advance which of these concepts is going to outperform next because in all trading, but especially when trading instruments as volatile as VIX products, there’s an element of randomness and the luck of the draw.

That means that the best way to ensure future success is to employ a bit of balance, to use a more holistic approach that draws from many different concepts.

The epic shift in the graph above has played out time and time again historically. That’s why you consistently find me staying very Zen about both a strategy that has performed very well recently and one that has performed poorly. Because the best strategy for the future, given what we know today, probably lies somewhere in between the two.

Click to see Volatility Made Simple’s own elegant solution to the VIX ETP puzzle.

Good Trading,
Volatility Made Simple