This is a test of a strategy from Ilya Kipnis of QuantStrat TradeR for trading VIX ETPs like XIV and VXX. Ilya provides a framework for testing the robustness of a given set of trading parameters. I encourage you to read Ilya’s piece, but that isn’t the subject of this post. Here I test the strategy that resulted from Ilya’s analysis (with a twist).
Ilya’s post found 3 different parameter values that looked promising. Here I’ve combined them into one single strategy. Strategy rules follow (read about test assumptions):
- After the close, calculate the ratio: VXV (3-month VIX) divided by VXMT (mid-term VIX).
- Calculate the 60-day, 125-day, and 150-day average of that ratio. These are for the three separate strategies that we will combine into one.
- For each strategy, when both the current VXV/VXMT ratio is below the average and the average is below 1, that strategy is short vol (XIV). When both the ratio is above the average and the average is above 1, that strategy is long vol (VXX).
- Average the signal from all three strategies. For example, 2 short vol and 1 cash signal would average out to a 2/3 position short vol.
- Execute that signal at the following day’s close using a market-on-close order. In other words, this strategy has a 1-day lag. We’ve touched briefly on 1-day lags previously. Hold until a change in position.
Because VXMT data is only available from 2008, we can’t test this strategy back to mid-2004 as readers are accustomed to. The strategy looks promising though despite the limited data, at the very least because some thought went into parameter selection.
Note from the equity curve above (and drawdown curve below) that the strength of the strategy has been in managing losses, and the strategy has tended to lag buy & hold when XIV has been particularly strong.
The strategy spends about 65% of all days with some position on. I should note that of those days, the vast majority (92%) are short the VIX, meaning an important mechanism for the success of this strategy is moving to cash when the ratio is above or below the average (as opposed to switching between long and short VIX).
I look forward to seeing how this strategy performs out-of-sample. Like most of the strategies we test on our blog, we’ll continue tracking this one for the benefit of subscribers.
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.
Volatility Made Simple