Test of Varan’s QLD-VXX-XIV Momentum Rotation Strategy

By reader request, this is a test of Varan’s QLD-VXX-XIV Momentum Rotation strategy.

The strategy trades monthly, allocating between the best two ETPs from the previous month based on simple risk-parity. Strategy results from 07/2004 follow. Read about test assumptions.

20140825.01

Strategy rules (slightly modified from Varan’s original):

  • On the last day of every month, go long at the close the two top performing ETPs from the preceding month from QLD (Ultra QQQ), VXX (long VIX) and XIV (inverse VIX).
  • Allocate between the two ETPs based on simple risk-parity, which is calculated by Varan to be: w1 = [s2/(s1+s2)] and w2 = (100% – w1), where w = the asset’s % allocation, and s = the standard deviation of daily returns over the preceding month.
  • Repeat at the close of each month.

20140825.02

Our backtest differs from Varan’s original test in two ways:

  • Varan’s test began in 2011. We’ve added over six years of additional data (1).
  • Varan’s test assumed trades were executed on the “first of every month”, but was unclear as to whether trades were executed at the open, close or some other point in between. For simplicity’s sake, we’ve assumed trades were executed at the close on the last day of the month, but note that there was little difference in performance executing trades at other times.

20140825.03

There are 3 very different “states” of Varan’s strategy:

  1. About 15% of the time, the strategy is holding VXX and XIV. These two ETPs will essentially cancel one another out (XIV is the daily inverse of VXX). These periods didn’t have a significant impact on performance.
  2. About 24% of the time, the strategy is long QLD and VXX. This pair will significantly reduce beta risk, and is mostly a play on VIX futures backwardation and/or negative VRP and/or a significant spike in volatility.
  3. The majority of the time (61%), the strategy is long QLD and XIV. This is a high beta, fully directional position that will also benefit from any contango and/or VRP in VIX futures.

In the years prior to when Varan’s test began in 2011, the strategy would have had difficulty keeping pace with most of the simple strategies we’ve tested previously on this blog.

A big thank you to Varan for posting this strategy and to readers for submitting it for testing.

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

Good Trading,
Volatility Made Simple


Wonk note: data prior to the launch of XIV and VXX has been simulated. We’re able to do this accurately using a combination of the indices and the futures data on which these ETPs are based. Read more about simulating data for VIX ETPs.

Posted in Strategy Backtests.