Seeking Alpha

Fred Piard's  Instablog

Fred Piard
Send Message
Author of 'The Lazy Fundamental Analyst' and 'Quantitative Investing' (4 and 5 stars at Amazon). I design portfolio strategies for clients and myself. I send weekly and monthly Newsletters. I hold a PhD in computer science, 2 MSc in software engineering and civil engineering, have worked 20... More
My company:
Ypa Finance
My blog:
Ypafi Global Household Index
My book:
2 Books With Real Alpha Inside
View Fred Piard's Instablogs on:
  • A New Money Management For My 3 VIX Models

    Disclaimer: This is a blog post for followers and subscribers, not a regular article. All readers are welcome. Past performance, real or simulated, is not a guarantee for the future. Trading is risky. I am not a RIA. This is a follow-up of my portfolio for informational purposes, not investment advice. I currently allocate about 5% of my main trading account to VIX trading strategies.

    A few points to remind what we are speaking of...

    • The 3 models are based on different logics and may behave differently depending on market conditions.
    • Model1 looks better when the VIX is low or in a whipsaw market (intra-week oscillations).
    • Model2 looks better in a mid-term trending market (S&P500 trending up about 1 month or more).
    • Model3 looks better in a short term trending market (S&P500 trending up or down about 1 week or more).
    • Model1 and Model3 may suffer badly when stocks and T-bonds fall together on the short term.
    • Model2 and Model3 may suffer badly in a brutal volatility spike
    • Model2 and Model3 have 1 position, Model1 has 2. Possible holdings are XIV, ZIV, TMF, VQT (or PHDG), cash.
    • Backtest period: since 1/1/2011 with this list, since 2009 with reverse short models using VXZ and VXX.
    • Model1 and Model2 were launched in Sept. 2013 in my personal account, and in Dec. 2013 in my VIX Trading Signals newsletter. Model3 was designed at the same time, but added in June 2014 in the newsletter.

    The models have done well in the first 7 months of 2014, but are in drawdown since August. The maximum drawdown limit has been broken from about -15% to about -18% for Model1 and Model2. Does it mean that my VIX trading should be stopped? In fact, after a closer look at the theoretical equity curves, only Model2 has broken its 2-year trend line. Model1 and Model3 still look good (however, the situation can change quickly).

    For months I have an intuition that the models are complementary, but I hadn't found out the way to combine them. Trading them depending on market conditions sounds too complicated. How to detect that market conditions have changed? This is just pushing the problem to another level. Should I invest in equal weight in the 3 models? It seems too risky when they are all in XIV. I also refused to do some «curve fitting» on the models with recent data. So, I came back to a basic principle in trading: everything is in money management.

    My idea is quite simple and based on 2 points:

    1. Combining the models to get the list of holdings, and equal-weighting the holdings - not the models. It means that if XIV is selected in more than one model, it is not over-weighted in the portfolio.
    2. Setting a maximum position size in percentage of the portfolio. It means that a part of the money allocated to VIX Trading may be in cash.

    The tables and charts below show simulated performances of the models separately, then combined in this way. First, I have combined Model1 and Model3, then the 3 models together. I have tried a maximum position size of 25%, 33.4%, 50%. Commission and slippage are modelled by a 0.2% cost on each transaction. The models were designed in summer 2013, so the performances last 14 months are out-of-sample.

    All combinations reduce the risk in terms of volatility and drawdown. The best risk-adjusted return is for Model1+Model3 with a maximum position size of 50%. My personal preference is for a lower drawdown with the 3 models and a maximum position size of 33.4%. It is likely the solution I will apply on my own account starting next week (I might even choose a max position of 25% for a lower risk).

    Data and charts: Portfolio123

    Model1:

    (click to enlarge)

    Model2:

    (click to enlarge)

    Model3:

    (click to enlarge)

    Model1+Model3, equal weighted holdings, no duplicated position, max position 50%:

    (click to enlarge)

    Model1+Model3, equal weighted holdings, no duplicated position, max position 33.4%:

    (click to enlarge)

    3 models mixed, equal weighted holdings, no duplicated position, max position 50%:

    (click to enlarge)

    3 models mixed, equal weighted holdings, no duplicated position, max position 33.4%:

    (click to enlarge)

    3 models mixed, equal weighted holdings, no duplicated position, max position 25%:

    (click to enlarge)

    Tags: XIV, ZIV, VXX, VXZ, VIX, Volatility, Bonds
    Nov 24 6:02 PM | Link | 2 Comments
  • Out-Of-Sample Test Of My Book's Strategies

    Last February my editor asked me if I would consider writing a second book after the good reviews of my first one «Quantitative Investing». Then I started «The Lazy Fundamental Analyst». Even if it was published in October 2014, it is a collection of 20 sector strategies on which I have worked in August and September 2013. In fact, I have designed this strategies for myself in case my usual tools are not available any more. A little bit paranoid? I think that being an independent investor is first being independent of the tools. So they are my back-up models, simple enough to be executed manually with a free screener like Finviz.

    I have just had a look at the performance of some of them last 12 months. This is an out-of-sample performance, or in plain English a backtest on a time interval after the period used for the design. Of course, past performance is never a guarantee, should it be real, simulated, or out-of-sample.

    All the following strategies have 10 holdings in the S&P 500 indices, are specialized by sector and rebalanced every 4 weeks. A 0.2% rate is accounted for trading costs. There is no hedge and no market timing here.

    Strategy

    1-year return

    Benchmark ETF

    Benchmark return

    Alpha

    SP500-Discretionary

    7.8%

    XLY

    7%

    0.8%

    SP500-Energy

    -21.8%

    XLE

    -1.2%

    -20.6%

    SP500-Financial

    17.6%

    XLF

    18.1%

    -0.5%

    SP500-Healthcare

    40.8%

    XLV

    28.7%

    12.1%

    SP500-Industrial

    27%

    XLI

    14.4%

    12.6%

    SP500-Materials

    16.8%

    XLB

    10.8%

    6.0%

    SP500-Staple

    23.4%

    XLP

    13.5%

    9.9%

    SP500-Technology

    30.5%

    XLK

    22.8%

    7.7%

    SP500-Utilities

    21.8%

    XLU

    20.5%

    1.3%

    Average

    18.2%

     

    14.9%

    3.4%

    Seven out of nine have a positive alpha, and there is only one negative return in Energy (and really ugly). The Energy strategy selects companies among the best Price-to-Book ratios in the sector, and they happened to have suffered the most with oil price: the strategy lost more than 30% since July.

    What is the cure when complete industries are falling? A top-down approach may do the job. It means adding a tactical allocation layer before picking stocks. For example keeping the 4 best sectors based on a momentum ranking, or eliminating the 2 worst ones. I hope to write a regular article about this in the next weeks.

    Tags: XLP, XLK, XLV, XLB, XLI
    Nov 05 8:51 AM | Link | Comment!
  • Facts About Ebola That You Must Know As An Investor, Traveler, Or Simple Citizen.

    About airlines

    There is an irrational fear erroneously focused on airlines, and some people are tempted by isolationism. History tells us that stopping commercial flights with Western Africa, or even in the entire world, would be of little use. During the 1918-1919 Spanish flu pandemic, a H1N1 virus killed 50 to 100 million people in a few months, including in the most remote parts of the world. It happened before the first commercial flight of a regular airline (in July 1919 between England and France). Fortunately, unlike the virus of 1918, ebolavirus is not airborne, its incubation is longer, and it is not contagious until the first symptoms. For readers still uninformed about the disease and its transmission, I strongly recommend to visit the World Health Organization Ebola page. One of the best informed article I have read on the current outbreak is an interview by Die Welt of a Washington based virus expert. Unfortunately it is in German. To summarize it, he expects the disease to peak around mid-January and then go down. He explains that it is especially lethal in Western Africa because of a weak healthcare system combined with cultural factors. He dismisses a similar scenario in developed countries and he is more concerned about the ignorance and hysteria surrounding the disease.

    I may add: if more airports implement passenger temperature scanning like in West Africa, New York and London, planes and boarding rooms may become the places with the lowest risk to catch any kind of infection.

    About survivors

    Everyone is waiting for a vaccine and cure from the pharma industry, but few people in the public know that the human body may be able to eliminate the virus. The best hope for each patient to win the race now is treating the symptoms, supporting the immune system with experimental drugs, and injecting survivors' antibodies. As always when the immune system is involved, the patient's previous medical condition, nutritional state, and stress are important. Survivors seem to be immune for several months and possibly more. It opens the possibility to train immune health workers.

    About the risk

    Coming back to the 1918 Spanish flu, a statistical study published in 2006 in The Lancet shows that the excess mortality rate varied over 30-fold across countries in inverse relation with the per capita income. Most of the difference can be explained by scarce health resources in poor countries, but nutrition may also be a significant factor. Guinea is one of the poorest countries in the world with a GDP per capita of $492, vs. $53.001 in the U.S. (source Wikipedia). The outbreak mortality rate is currently about 50% in Western Africa. Ebola is not Spanish flu: extrapolating the 30-fold factor would be speculation. But it helps understand that in a bad scenario, the mortality rate should be much lower in developed countries, even in a situation with no vaccine and no cure (there was no in 1918). Moreover, the pharmaceutical and medical authorities have said for years that supplements have no proven effects in treating diseases, but things change. Now the very prudent Harvard Medical School admits that some vitamins and minerals may help boost the immune system against viral infection like flu. There is no study in the case of Ebola, but any chance to make a small change in mortality rate can make a big difference.

    Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it. The author has no business relationship with any company whose stock is mentioned in this article.

    Tags: demographics
    Oct 15 12:01 PM | Link | 4 Comments
Full index of posts »
Latest Followers

StockTalks

  • urlz P/L in Global Household trading model (10 stocks): 1 loss $BSX -0.7%, 9 gains btw +1% and +14% $LH,$MNK,$MYL,$STJ,$CPB,$K,$LO,$SJM,$RAI
    about 1 hour ago
  • How Much Is The Biotech Industry Stretched? http://seekingalpha.com/a/1m8hr
    about 5 hours ago
  • Headache tomorrow for oil stocks
    1 day ago
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.