Studies - Posted on Sunday, September 5, 2010, 11:59 PM GMT +1

3 Comments


Sep Sunday 5

Williams’ VIX Fix (WVF)

Most recently Michael Stokes at MarketSci and MINDMONEYMARKET (I don’t know who is behind the blog) had – as always – some very inspiring postings about the Williams’ VIX Fix (WVF) (here and here). William’s VIX Fix is a synthetic VIX (CBOE Volatility Index) calculation which can be used in any market to mimic the performance (but not the quotes) of the well-known volatility index (the WVF is not based on option’s implied volatility but derived from historical and intraday prices only).

William’s original formula:

WVF = [Highest (Close,22) - Low) / (Highest(Close,22)] * 100

Michael and MINDMONEYMARKET already showed that at least with respect to the S&P 500, relatively high WVF readings were frequently (means with a probability significantly above the respective at-any-time probability) followed by a higher close on the then following session. Buying into high WVF readings (when the delta between the current intraday low and the highest close of the previous 22-trading days) is at positive extremes is therefore buying into strength, not betting on a short-term mean reversion tendency.

But with respect to the WFV‘s usefulness for any kind of trading strategy, Michael noted two shortcomings as well:

(1) The WMF‘s dependency on intraday data (the ‘Low‘), and (2) the WVF‘s potential lack of robustness due to the fact that less extreme WVF readings might significantly (negatively) impact a potential trading strategy. Both bullet points incited me to check if – and how to – slight adjustements to the formula cited above could solve these issues.

Table I below shows the SPY‘s historical performance assumed one went long on close of a session when the WVF closed

  • Strat. #1: among the top 4 of all readings over the previous 22 trading days,
  • Strat. #2: among the top 3 of all readings over the previous 22 trading days,
  • Strat. #3: among the top 2 of all readings over the previous 22 trading days,
  • Strat. #4: as the highest of all readings over the previous 22 trading days,

otherwise no position is taken (move to cash). The 22-trading day period is slightly deviating from what Michael and MINDMONEYMARKET used, but matches the one month period William’s chose for his original formula (see above).

Interesting to note that by increasing the WVF entry level (and therefore lowering the respective number of occurrences and time in market), compounded returns remain almost unchanged due to the fact that going long on close of a sessions where the WVF closed among the top 2 of all readings over the previous 22 trading days (Strat. #3 ) showed the highest median trade, profit factor and distribution of returns and the lowest drawdown among all entry levels. Lowering the entry level – at least with respect to the set of parameters utilized for this evaluation – wouldn’t turn a profitable strategy into a losing one, but would increase risk and expenditures (maximum drawdown, time in market, transaction costs …) with no additional reward.

The second question was if utilizing the intraday low as part of the original formula is a must, or if substituting the intraday low by the daily close (therefore relying on daily closing prices only) would negatively impact the formula’s usefulness and quality of forecast. Table II below shows the SPY‘s historical performance assumed one went long on close of a session when a ‘redefinedWVF closed among the top 2 of all readings over the previous 22-trading days (previous strat. #3), with

  • Strat. #1: WVF = [Highest (Close,22) - High) / (Highest(Close,22)] * 100,
  • Strat. #2: WVF = [Highest (Close,22) - Low) / (Highest(Close,22)] * 100,
  • Strat. #3: WVF = [Highest (Close,22) - Close) / (Highest(Close,22)] * 100,

otherwise no position is taken (move to cash). Strategy #2 represents William’s original formula.

And now going into one more extreme (shortened time frame):  Table III below shows the SPY‘s historical performance assumed one went long on close of a session when a ‘redefinedWVF closed among the top 2 of all readings over the previous 10-trading days (instead of 22-trading days), with again

  • Strat. #1: WVF = [Highest (Close,22) - High) / (Highest(Close,22)] * 100,
  • Strat. #2: WVF = [Highest (Close,22) - Low) / (Highest(Close,22)] * 100,
  • Strat. #3: WVF = [Highest (Close,22) - Close) / (Highest(Close,22)] * 100,

otherwise no position is taken (move to cash). Strategy #2 represents William’s original formula.

Interesting to note that at least with respect to utilizing the SPY for the time frame since 01/01/1990, relying on closing prices alone (no intraday high or low) would (partly significantly) improve the overall profitability (compound returns, probability of a higher close, median trade, distribution of returns), by having to accept a slightly larger drawdown (but significantly smaller number of maximum sessions in a drawdown). For shorter periods (e.g. the previous 10-trading days instead of the previous 22-trading days) using the intraday ‘High‘ instead of William’s intraday ‘Low’ would’ve worked best the majority of the time (see equity curve above, blue line), out-performing the ‘Low‘ in the original formula by a wide margin.

______________________

And in order to ensure that utilizing closing prices only instead of using intraday low or high for the computation of the WVF may be the better choice in principle and not only applicable for the SPY, I performed the same stats for the QQQQ (Nasdaq 100) as well (see table IV below, data since 01/01/2000).

It is again strategy #3, utilizing closing prices only instead of intraday data, which provides the highest rate of return, profit factor, median trade and distribution of returns.

_______________________________

Summary: Utilizing percentage rankings, and at the extreme (high) end of the then recent previous x readings, William’s VIX Fix seems to be a reliable (with respect to it’s quality of forecast) and very robust indicator forecasting the bullish side of the market (unfortunately WVF readings at the low end of the then recent previous x readings are not able to forecast weakness in the markets by any kind of statistically significance). But using daily closing prices instead of the intraday low may not only simplify the handling and computation but will probably improve the quality of forecast (with respect to probabilites and odds) as well.

Successful trading,
Frank

xx

Disclaimer: No position in the securities mentioned in this post at time of writing.

Remarks: Due to their conceptual scope – and if not explicitely stated otherwise -, all models/setups/strategies do not account for slippage, fees and transaction costs, do not account for return on cash, do not use position sizing (e.g. Kelly, optimal f) – they’re always ‘all in-, do not use leverage (e.g. leveraged ETFs) - but a marginable account is mandatory -, do not utilize any kind of abnormal market filter (e.g. during market phases with extremely elevated volatility) , do not use intraday buy/sell stops (end-of-day prices only), and models/setups/strategies are not ‘adaptive‘ (do not adjust to the ongoing changes in market conditions like bull and bear markets).

________________________________

If you might want to be instantly notified about what’s happening in the markets and at TRADING THE ODDS, I encourage you to subscribe to my RSS Feed or Email Feed, and (or) follow me on Twitter.

xx

The information on this site is provided for statistical and informational purposes only. Nothing herein should be interpreted or regarded as personalized investment advice or to state or imply that past results are an indication of future performance. The author of this website is not a licensed financial advisor and will not accept liability for any loss or damage, including without limitation to, any loss of profit, which may arise directly or indirectly from use of or reliance on the content of this website(s). Under no circumstances does this information represent an advice or recommendation to buy, sell or hold any security.

Add to Technorati Favorites


p5rn7vb

Comments (3)

 

  1. [...] Williams’ VIX Fix? Neither had I, until this weekend.  Here’s the description by Frank of Trading the Odds. William’s VIX Fix is a synthetic VIX (CBOE Volatility Index) calculation which can be used in any [...]

  2. Toptick says:

    Great work. Thanks!

  3. [...] several of these have posted on the "William's VIX Fix" (hereafter wvf): marketsci, trading the odds, mindmoneymarkets. The wvf is intended to be a synthetic VIX calculation, derived by Larry Williams [...]

Leave a Reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>