The Best Historical Data for Backtesting VIX Trading Strategies

I often get asked for the best source of historical data to backtest strategies for trading VIX ETPs like VXX and XIV. These products can be accurately estimated back to early 2004, nearly 5 years before the launch of the first VIX ETP.

In my own research I use the formal index on which these ETPs are based, called the “S&P 500 Short-Term (or Mid-Term) Futures Index TR”, but this data is expensive to acquire and probably unnecessary for retail traders.

A very close approximation can be had by manually recreating that index using the right mix of individual VIX futures contracts. This type of data is available from a number of sources online, but I expect that the very best of the bunch is from Vance Harwood of Six Figure Investing.

Vance sells estimated close data for a number of popular ETPs here, and estimated open/high/low values here, all at a very reasonable cost.

As anyone who has followed Vance’s work knows, he has an incredibly detailed eye for the minutiae of the VIX and VIX products, and his historical data reflects that.

If you’re new to VIX trading research, I highly recommend starting with a good dataset like Vance’s. Note that I’m in no way being compensated for this post – just shining a light on the good guys to help future VIX nerds get pointed in the right direction.

Good Trading,
Volatility Made Simple

Required Reading: How Does the VIX Index Work?

For those new to trading the VIX and volatility, the always outstanding Six Figure Investing has written a very nice primer on the VIX index. Six Figure’s primers are always highly recommended.

Just to clarify the relationship between the VIX index and what we do:

We trade VIX ETPs (like XIV and VXX). VIX ETPs track a specific mix of VIX futures. VIX futures target the future value of the VIX index. In summary: VIX index -> VIX futures -> VIX ETPs.


On a less pleasant note, the short-term VIX ETP XIV was down -3.5% on the day. As I’ve written about previously, these type of daily moves, while huge for most assets, are business as usual with VIX products – the inherent cost of capturing big returns.

Since mid-2004, XIV would have lost in excess of -3.5% about 27 times per year, or more than twice per month on average. Business as usual.

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.

Required Reading: How Does XIV Work?

Readers know that I’m a proponent of trading long XIV (and sometimes ZIV) as opposed to short VXX or VXZ. As evidenced by the number of emails I receive asking why that is, I owe a post on the subject.

Consider that added to the long to do list, but in the meantime, here is a bit of required reading for folks new to XIV from the always outstanding Six Figure Investing: How Does XIV Work?

Good stuff, and well worth the read.

Good Trading,
Volatility Made Simple

New Subscriber Feature: Tracking Blogged Strategies

When one of the key strategies we test on our blog signals a change, 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.

But keeping track of all of these strategies can be difficult, especially as we continue to blog and add new studies, so we’ve launched a new feature for subscribers: a summary of all blogged strategies.

We sort the strategies from most to least effective historically, based on Sharpe Ratio, UPI, and recent returns, then color code them by current position, and denote when that position changes. Yesterday’s summary would look as follows. Note the greater caution being shown by the better-performing strategies.

Strategy
Position at close on 04/21/14
50% Inverse Vol (XIV/ZIV)
100% Inverse Vol (XIV/ZIV)
100% Long ZIV
100% Cash
100% Long Vol (VXX/VXZ)
100% Cash
100% Inverse Vol (XIV/ZIV)
100% Inverse Vol (XIV/ZIV)
100% Inverse Vol (XIV/ZIV)
100% Inverse Vol (XIV/ZIV)

The report is not perfect. For example, it’s too heavily weighted at the moment in strategies that key off of the futures term-structure (ex. comparing first and second month VIX futures), but that will change over time as our blog fills out and I continue to tick blog posts off of my long to-do list.

To receive daily updates to our strategy (as well as this handy dandy summary of all blogged strategies), subscribe for as little as $1 a day.

Good Trading,
Volatility Made Simple

The Ulcer Performance Index

The statistics I show here at Volatility Made Simple almost always include one of my favorites: the “Ulcer Performance Index” (UPI).

The UPI is like a Sharpe ratio that views risk in terms of the length and severity of drawdowns, rather than in terms of volatility. I show max drawdown as well, but in actuality, I think the stat is overused (and often abused). Loss at a single point in time (max drawdown) is far less important than how a strategy manages losses over time (UPI).

Flirting with Models has a nice write up on the subject that I highly recommend. I particularly liked the “car driving” analogy (and have appropriated it for future use). For those just interested in the math, see Wikipedia.

Note that Flirting with Models is talking about the Ulcer Index (UI), which is simply the denominator in the Ulcer Performance Index (UPI) calculation. In other words…

UPI = [(annualized return – risk free rate) / UI]

(h/t The Whole Street)

Good Trading,
Volatility Made Simple