Seeking Alpha

Fred Piard's  Instablog

Fred Piard
Send Message
$$$ NEW BOOK: The Lazy Fundamental Analyst $$$ - I design systematic strategies for clients and myself. I trade stocks and ETPs. For individual investors, I publish Newsletters mirroring my portfolios. For professionals, I set up personalized services based on investment themes (surprises,... More
My company:
Ypa Finance
My blog:
Ypafi Global Household Index
My book:
2 Books With True Pieces of Alpha
  • Why I Am Still Trading XIV And ZIV. 6 comments
    Apr 11, 2014 12:06 PM | about stocks: XIV, ZIV, VXX, VXZ, TMF, PHDG, VQT

    Recently, several articles have been published here and elsewhere claiming that traders should stop shorting volatility now. Obviously, 2014 didn't start in a so impressive way as 2013, however my thoughts are not so negative about that and I would like to share my recent experience here.

    I trade two weekly trading models on XIV and ZIV (long side only). They rely on a combination of rules using momentum, relative strength, VIX index, and inter-market behavior. Basically, they are trend-following models, not "reversal to the mean" ones. The design process was partly described in this post. According to backtests since 2011 (and even since 2009 translating them in shorting VXX and VXZ) both of them are quite similar in drawdown, volatility and return. Nevertheless they have different philosophies:

    - Model1 switches between XIV, ZIV and cash. It is in the market "as often as possible", most of the time hedged by a leveraged bond ETF (NYSEARCA:TMF). This hedge is a drag for the total return, but it gives a significantly better risk-adjusted performance (Sharpe and Sortino ratios) by cutting drastically volatility and drawdowns.

    - Model2 trades XIV and uses a Veqtor Index ETP (VQT or its equivalent PHDG) as a waiting position. The Veqtor index embeds a daily quantitative strategy with a long exposition to volatility in strong market downturns. Model2 is in waiting mode "as often as possible" and gets into XIV only when the probability of making a profit is high according to historical data.

    To summarize, these models have different tactics: a loose timing with hedging on the one hand, and a restrictive timing combined with an embedded external strategy on the other hand.

    Here are the backtested returns and max drawdowns year by year since 2011, compared with XIV and SPY:

     

    2011

    2012

    2013

    2014 YTD*

     

    Return

    maxDD

    Return

    maxDD

    Return

    maxDD

    Return

    Annualized

    maxDD

    Model 1

    71.8%

    -9.6%

    69.4%

    -8.3%

    36.3%

    -14.3%

    9.06%

    37.7%

    -4%

    Model 2

    91.0%

    -14.8%

    122.9%

    -13.8%

    72.6%

    -10.5%

    -6.7%

    -22.4%

    -12.7%

    XIV

    -44.8%

    -74.6%

    167.2%

    -37.2%

    86.1%

    -27.7%

    -10.1%

    -32.4%

    -27.4%

    SPY

    2.5%

    -20.2%

    16.1%

    -9.8%

    29.5%

    -5.5%

    0%

    0%

    -5.4%

    *4/10/2014 on close

    Positions are taken on weekly open until the next weekly open. A rate of 0.2% is accounted for trading costs.

    As you can note, the idea of the models is not necessarily to outperform XIV, but to give a protection against drawdowns and offer a better risk-adjusted performance.

    The year started with a bad trade of Model2 in the end of January, and it may happen again. However I remain reasonably optimistic on the long term. Here are the reasons why:

    - Both models did very well in 2011, whereas XIV fell like a brick.

    - Both models outperform XIV year-to-date, with a much lower volatility and drawdown.

    - The equal-weight combination of both models is slightly in the green YTD, whereas SPY is flat and XIV is down 10%.

    I am positively impressed by the performance of Model1 in the current tough market conditions for short-volatility traders, especially for a model that trades only once a week. Its annualized return YTD stays similar to 2013. As it seems more resistant in a changing market, I might consider give it a higher weight relatively to Model2 in the future.

    As a conclusion, shorting volatility doesn't seem to be dead for 2014. Traders with either good discretionary skills, or a good model, may be able to navigate in this market. At least I hope so. Studying past data is not a guarantee for the future, but it is the best way to plan it.

    Caution note: volatility trading is closer to scientific gambling than to investing. Depending on an investor's risk appetite, it might be suitable as a satellite portfolio with a small percentage of capital.

    Disclosure: I am long ZIV, TMF, PHDG.

    Stocks: XIV, ZIV, VXX, VXZ, TMF, PHDG, VQT
Back To Fred Piard's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (6)
Track new comments
  • jeezuz30
    , contributor
    Comments (325) | Send Message
     
    Thanks for the article. Just curious with your knowledge why not go both long and short volatility, depending on direction.

     

    Seems like this year is starting out good with shorts bursts or volatility, eg from around jan 27th through feb 3rd and what looks like another little bout starting a few days ago.

     

    Like you mentioned , somebody that can ride the waves is bound to make profits, why restrict yourself only to the short side?
    11 Apr, 04:17 PM Reply Like
  • Fred Piard
    , contributor
    Comments (605) | Send Message
     
    Author’s reply » First because it is the direction of the contango bias on VIX futures 80% of the time (article here: http://bit.ly/1ghHuZs). Second because I prefer trading on a weekly timeframe. If I trade long VXX with weekly setups, probabilities are stacked against me. I have a "theoretically good" daily model long VXX but I don't trade it: highly volatile, not many signals, not my style.
    11 Apr, 05:39 PM Reply Like
  • jeezuz30
    , contributor
    Comments (325) | Send Message
     
    Fair enough, indeed a weekly model would only work on the short side.
    1 May, 01:32 PM Reply Like
  • Augustus
    , contributor
    Comments (1957) | Send Message
     
    glad to read your update and assessment. Thanks.
    12 Apr, 01:09 AM Reply Like
  • RideTheTide
    , contributor
    Comments (2) | Send Message
     
    VQT/PHDG price action generally correlates to the VIX for spikes 20 and under -- only kicking in its true hedge value for larger spikes (e.g. 2011 spike). So if you use them as a holding position for XIV entry on Model 2 wouldn't you be taking haircuts liquidating those positions at their recent lows to initiate a new position in XIV? Or does the model also time entry positions to VQT/PHDG to capture entry at a discount?
    13 Apr, 08:20 AM Reply Like
  • Fred Piard
    , contributor
    Comments (605) | Send Message
     
    Author’s reply » "VQT/PHDG price action generally correlates to the VIX for spikes 20 and under"
    => this is more complicated than that. You can find a methodology document about the Veqtor index on http://spindices.com

     

    The model 2 was originally designed without VQT/PHDG (waiting mode in cash). Simulations shows that putting the idle cash in VQT brings an additional return and no additional risk since 2011. As you noted, it gives a long exposure to volatility in significant downturns (in my profile you can find two more detailed articles about VQT/PHDG).
    About possible haircuts: taking losses is part of the game in trading, this is not value investing and even less "buy and hold". It is about the probability of gain and the average gain/average loss ratio.
    14 Apr, 02:35 PM Reply Like
Full index of posts »
Latest Followers

StockTalks

More »

Latest Comments


Posts by Themes
Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.