Playing the Short Side on SPX
I’ve been experimenting these last few days on fixing up my “optimal” contango ratio to enter or exit short volatility. This came about because of someone’s response to my post in Quantopian; essentially, given that I’m checking the ratio between the VIX Spot and the Front month futures’ settlement, it’s to be expected to converge to 1 as we approach Futures maturity.
I’ve been meaning to think of a way to resolve this and ended up just doing something similar to the construction of the ETF VXX where I take a day-weighted average of the VIX/F1 and F1/F2 ratio (something similar in concept is in my earlier post). This new standardized ratio works out quite well and reduces the drawdown by quite a fair amount where if Ratio < 0.95, short volatility, otherwise move to cash.
Below are two back-test comparisons for the time-frame 12/01/2010 to 01/20/2017.
Long XIV
Short VXX
With shorting VXX, I’ve ran further back-testing analysis since I can actually go back a little farther in time to 02/02/2009. I’ve detailed a summary of the returns below.
|
|
| Stress Events | Mean | Min | Max |
|---|---|---|---|
| US downgrade/European Debt Crisis | 0 | 0 | 0 |
| Fukushima | 0.0054 | -0.0173 | 0.0482 |
| EZB IR Event | 0.0015 | -0.0745 | 0.1064 |
| 2009Q1 | 0 | 0 | 0 |
| 2009Q2 | -0.0015 | -0.0533 | 0.0142 |
| Flash Crash | 0 | 0 | 0 |
| Apr-14 | -0.0022 | -0.0549 | 0.0319 |
| Oct-14 | -0.0009 | -0.013 | 0.0029 |
| Fall 2015 | -0.0019 | -0.0444 | 0.012 |
| GFC Crash | 0 | 0 | 0 |
| Recovery | 0.0031 | -0.1242 | 0.1251 |
| New Normal | 0.002 | -0.0911 | 0.1061 |
Related posts
Amazing stuff, Fang! I’ve taken your algorithm and implemented it into some of my portfolio. Thanks for being kind enough to share with the world. 🙂
No problem! I’m just building off of someone else’ research too; collaboration makes everyone better =D