The Mojito


I like girls like my asset classes. Bubbly.


What do you mean you don't know whether whether we should take the business or not?

All money is green.


I like my cocktails like my money. Green!



That was the CEO sitting behind a younger me.

He didn't have an office to himself when he came over to Ireland, so he sat behind me.

Start-ups breed characters - good and bad - that edginess is what you need when you're up against corporate behemoths.

In any case, in my ever widening search for edge, Conor pointed out a bunch of papers on VIX ETN trading strategies (1,2) after my post on a similar strategy.

Understanding term structure movements is the blackest of arts, whenever I would contact hedge funds, they'd usually interrogate me about the quality of our Swaption data for example.

Not because they wanted a risk system which was founded upon good data, but because they were pitching themselves to investors as macro strategy gurus, but they could never get their hands on decent data and were eager to find some.

Chrilly Donninger (2) works through several VIX ETN strategies in his research note and comes up with the Aggressive Mojito strategy.

The Mojito strategy uses a ratio of the spot VIX over a 3 month future of the Vix (VXV) to predict future contango or backwardation in the term structure and then bets on the short to medium term structure.

It is very similar to the strategy in the previous post.

Logic for the Lazy Backtesting IDE is here.

It worked exceedingly well up until he and Carol Alexander published their papers. Coincidence?

But the real issue here of course is a lack of data, we need lots more data to judge its veracity properly.

The key takeaway is the Mojito is another strategy which converts volatility into a pretty bubbly asset class orthogonal to equity.

The Sharpe is 1 vs 0.86 of my previous strategy. Not bad at all!

Skew is not fantastic, but OK.

How orthogonal is this to the S&P 500?

Adding the previous strategy resulted in a 25% bump from 0.6 to 0.75.

The original Vix strategy has a Sharpe of 0.86, so a truly orthogonal strategy to the S&P 500 would result in a Sharpe of 1.46 or so, right?

By adding them to the same portfolio, we effectively halved this mythical orthogonal 1.46 target to 0.75. Frustrating.

How about we weight the portions devoted to the Mojito and the S&P 500 equally?

We would expect a orthogonal Sharpe target of around 1.6, right?

What we actually come up with, is a Sharpe 1.458.

Achieving over 90% of the target.

Enough Vixyness is left intact with the Mojito that over the last 6 years it's provided that little pep when the S&P 500 has flagged.

I am naming this potent mix the 'Angry Mosquito' strategy after all my Singapore mozzy friends.

Logic for the Lazy Backtesting IDE is here.

Again, diversification is the only good deal in finance - and volatility as an asset class can provide this in spades.

[Thanks again to Conor. Visit his blog , it's great!]