Seeking Alpha

Fred Piard's  Instablog

Fred Piard
Send Message
I design portfolio strategies for clients and myself. Author of 'The Lazy Fundamental Analyst' and 'Quantitative Investing' (5 stars at Amazon). Editor of the Ypa Finance newsletters series: Market Neutral Portfolio, Growth Portfolio, Dividend Portfolio. Ypa Market Neutral is a DIY hedge fund... More
My company:
Ypa Finance
My blog:
Ypafi Global Household Index
My book:
2 Books With Real Alpha Inside
  • Results Of My VIX ETPs Models, January To June 2014 3 comments
    Jun 3, 2014 11:30 AM | about stocks: XIV, ZIV, TMF, VXX, VXZ

    I invest in 2 models using inverse VIX ETPs since the end of Q3 2013. They may take positions in XIV, ZIV, and also in a leveraged bond ETF (NYSEARCA:TMF) and in the VEQTOR index (VQT or its equivalent PHDG). This is not intensive trading: I rebalance them once a week.

    They have different logics:

    - Model 1 switches between XIV, ZIV and cash. When it holds XIV or ZIV, it takes an equal-weight position in TMF. It is in the market "as often as possible": it means that it uses permissive timing rules. Most of the gain comes from XIV. TMF acts as a hedge to reduce the risk. In fact, the simulated historical return is better without TMF, at the price of a higher volatility and deeper drawdowns. The leveraged bond position improves significantly the risk-adjusted performance.

    - Model 2 switches between XIV and VQT (or PHDG). In fact, the Veqtor ETP is a waiting mode between two XIV trades. The waiting time can also be spent in cash: the core of the model is XIV, the Veqtor index brings only a small additional bias. Veqtor may capture a part of the volatility spike in a deep market downturn, as it did in 2011. It is a relatively safe way to wait for the next trade with a dynamic exposure to stocks and volatility. Model 2 has more restrictive timing rules: it is in XIV only when the odds are strongly favorable according to the rules.

    Because of ETPs inception dates, both models have been backtested since 1/1/2011 with their normal holdings, and since 4/1/2009 with inverted rules and short positions in VXX, VXZ and TBT. The period doesn't include a bear market. I have simulated simplified models since 2004 using synthetic prices. The bias looks positive, but it is not a backtest.

    To make things short, the historical annualized return of Model1 and Model2 combined are about 60% on the period of simulation, with a max Drawdown about 15%. Quite attractive, but past performance is never a guarantee for the future, especially when it is not evaluated in all market conditions. On the other hand, it would be frustrating to design such models and just look at them on the paper. I bet about 7% of my main trading account on this. The rest is in quantitative strategies on ETFs and S&P 500 stocks (full YTD results here), and cash. I use 10% to 20% trailing stops on XIV (there are no stops in sims).

    Here are the real performances year-to-date (1/1/2014 - 6/5/2014):


    Total return

    Annualized return

    Max Drawdown

    Sharpe ratio


    Model 1






    Model 2






    Model1 was especially robust in this whipsaw market.

    Models in equal weight returned 12%, with an 8% drawdown. In the same period, XIV had a better return: 20%, at the price of a scary -27% drawdown and 46% in volatility.

    So yes, I lagged XIV in the 5 first months of 2014. But I prefer a better risk-adjusted performance than a better total return. I sleep much better than holding XIV all the time, and it takes only a minute per week. As I write this, both models are in waiting mode (cash and Veqtor).

    I may add a 3rd XIV/ZIV weekly model in my toolbox: it also uses TMF, but not for hedging: it is a sophisticated switching model with timing rules based on inter-market behavior. I kept it in observation for more than 6 months to figure out how it would recover from a steep drawdown last summer.

    Here is the inverted rules backtest of Model3 with short positions in VXX, VXZ and TBT since 2009:

    (click to enlarge)

    The regular Model3 backtest (with XIV, ZIV and TMF) shows a 144% annualized return since 1/1/2011, with a 34% drawdown. This is the most impressive in total return, but also the most volatile. In fact, it is similar to Model1 and Model2 in terms of risk-adjusted performance. Adding a new model is interesting when it has a different logics, so it may behave differently in specific (and unknown) market conditions. Model3 is up 20% YTD on paper: I have no money on it yet.

    Disclosure: I am long PHDG.

    Stocks: XIV, ZIV, TMF, VXX, VXZ
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 (3)
Track new comments
  • Edmund Shing
    , contributor
    Comments (31) | Send Message
    Very nice strategies Fred, I believe we share a book editor in Stephen Eckett at HH! I was wondering about your 3rd strategy: would this involve mixing up XIV, ZIV and TMF (instead of cash)? In which case, you would currently be invested in TMF? Edmund
    3 Jun, 11:59 AM Reply Like
  • Fred Piard
    , contributor
    Comments (641) | Send Message
    Author’s reply » Thanks Edmund. About HH: yes, I'm waiting for a feedback on my second book's manuscript (no VIX trading inside, neither in the 1st book). I don't invest yet in Model3. Some checking to do about logical interactions with other models before. It would be in ZIV this week.
    3 Jun, 12:52 PM Reply Like
  • Fred Piard
    , contributor
    Comments (641) | Send Message
    Author’s reply » Updated 6/5 after close.
    6 Jun, 11:08 AM Reply Like
Full index of posts »
Latest Followers


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.