Seeking Alpha

varan's  Instablog

varan
Send Message
Individual investor.
  • Paired Switching for Q1 2012 14 comments
    Jan 6, 2012 2:15 PM | about stocks: VTI, EFA, EEM, IEZ, GLD, USO, TLT
    During 2012, I will be following the performance of the paired switching strategy for  VTI, EFA, EEM, IEZ, USO, and GLD, each paired with TLT. The method entails the comparison of  the performance of the ETF with that of TLT during the prior twelve weeks and investment in the better performing ETF, and repeating the cycle every thirteen weeks.

    Based on the performance during the period 10/7/2011 thru 12/30/2011, the strategy leads to the following selections in which to invest at the close of market today, 1/6/2012.

    PairJanuary 6, 2012 Selection
    VTI/TLTVTI
    EFA/TLTTLT
    EEM/TLTEEM
    IEZ/TLTIEZ
    GLD/TLTTLT
    USO/TLTUSO


    Clearly, this is not investment advice, but just my attempt to follow this strategy which seems to do well in back testing.

Back To varan'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 (14)
Track new comments
  • alchem
    , contributor
    Comments (10) | Send Message
     
    Please comment re the above:
    If USO favored over TLT, then why purchase TLT if favored in another pair, e.g. gold vs TLT and EFT vs TLT? Or counter argument is do not purchase USO if TLT favored twice?

     

    How would you balance the portfolio with TLT (%) given the above results?

     

    thanks
    7 Jan 2012, 06:10 PM Reply Like
  • varan
    , contributor
    Comments (3613) | Send Message
     
    Author’s reply » That's a good question. The pairs are traded independently, so this situation arises often. The result is that there is greater chance of USO doing better than TLT. Simultaneously TLT is likely to do better than GLD.
    7 Jan 2012, 07:25 PM Reply Like
  • Demi9OD
    , contributor
    Comments (5) | Send Message
     
    What about diversifying more by using a combination of the buy and hold weighting with the paired switching?

     

    Remove TLT from the buy and hold weighting and you are left with something like this:
    VTI 17%
    EFA 16%
    EEM 17%
    IEZ 15%
    USO 17%
    GLD 18%

     

    Substitute TLT for each side of the pair which it outperformed in the ranking period, and you are left with:
    VTI 17%
    EEM 17%
    IEZ 15%
    USO 17%
    TLT 34%

     

    Have you had a chance to re-balance the buy and hold portfolio for this year? How is this done using historical data?
    7 Jan 2012, 11:11 PM Reply Like
  • varan
    , contributor
    Comments (3613) | Send Message
     
    Author’s reply » Other variations may be possible, but I used the simplest one. Just run the simulation for the six pairs, and count the total number of days during which each ETF was held. This allows you to compute the percentage of the time during which a position was held in each ETF (simply divide the number of days for each ETF by the sum of these numbers). I take these percentages to be the weights for the buy and hold portfolio.
    8 Jan 2012, 12:51 AM Reply Like
  • michael.cave
    , contributor
    Comments (4) | Send Message
     
    Any updates on the 2Q performance of this strategy?
    1 Jul 2012, 03:13 PM Reply Like
  • varan
    , contributor
    Comments (3613) | Send Message
     
    Author’s reply » Not too good I am afraid.

     

    Week of 1/3/2012 to 4/2/2012 VTI, USO, IYZ , EEM, TLT, TLT

     

    Return 3.32%

     

    Week of 4/2/2012 to date EEM, EFA, VTI, GLD,IYZ,USO -5.2%

     

    Net YTD -2.12%

     

    The ranking for the last 12 weeks suggests to go full bore with TLT (who would have thought that in this low interest rate environment, TLT will beat all of these?).

     

    Please note that this is not investment advice in any shape or form. I am just trying to assess how this strategy performs.

     

    Given the caveat, and the fact that the backtesting for 2007-2011 yields very good result, I will give it a few more quarters without trying to modify it.

     

    Thanks a lot for your interest.
    1 Jul 2012, 03:57 PM Reply Like
  • Market Map
    , contributor
    Comments (223) | Send Message
     
    In 3rd qtrs. , VUSTX has beaten the SPX in aggregate since 1991 ...

     

    http://tinyurl.com/7tk...

     

    1 Jul 2012, 04:12 PM Reply Like
  • varan
    , contributor
    Comments (3613) | Send Message
     
    Author’s reply » thanks.

     

    the problem for designing this type of strategies is that with back testing you get immediate results, but real time performance takes too long to evaluate.
    1 Jul 2012, 05:01 PM Reply Like
  • michael.cave
    , contributor
    Comments (4) | Send Message
     
    Varan... Looks like another disappointing quarter for this strategy. I really like the simplicity of the strategy and plan to keep tracking it. Any thoughts on modifications?
    30 Sep 2012, 03:28 PM Reply Like
  • varan
    , contributor
    Comments (3613) | Send Message
     
    Author’s reply » Sorry, this year has indeed been quite dismal, although even including the 2012 returns, the annualized return since 2007 is still over 20%.

     

    For the sake of keeping thing honest I should let it play it out till the end of the year, although the time periods specified in this strategy for ETFs

     

    http://bit.ly/Vzg20t

     

    (invest the top six for 15 weeks based on the prior eight weeks of performance) has done quite well in all the years since 2007 (13.3% this year).

     

    For what it's worth the ETF strategy of the linked post has returned 19.8%, and the strategy for the select funds in one of my other instablogs about 20%. I mention this not to claim any special insight or expertise, as, like everyone else, I am exploring ways to enhance the returns of my portfolio, but just to illustrate that these kind of strategies do tend to work most of the time.
    30 Sep 2012, 04:21 PM Reply Like
  • michael.cave
    , contributor
    Comments (4) | Send Message
     
    Have you tried combining the paired switching with the select funds strategy?
    2 Nov 2012, 07:24 PM Reply Like
  • drftr
    , contributor
    Comments (117) | Send Message
     
    Varan,

     

    Did you do any experiments with pairing for instance SPY with its inverse cousin SH? At least theoretically this should work great. The only situation where results could be poor would be in sideways markets I think.

     

    drftr
    30 Aug, 09:41 PM Reply Like
  • varan
    , contributor
    Comments (3613) | Send Message
     
    Author’s reply » That does not work well.

     

    My conjecture is that the correlation should not be perfect -1 for this methodology to work.
    30 Aug, 10:40 PM Reply Like
  • drftr
    , contributor
    Comments (117) | Send Message
     
    Hmm... That's interesting... It may be because for instance you start buying SPY when it's above +1% and SH when it's below -1%, while most of the time the index will be exactly in the +1%/-1% range, so when you're not in the market. That's food for thought. Well, for me. You've obviously been through that phase already.

     

    Thanks for your time and efforts to teach me/us Varan!

     

    drftr
    1 Sep, 10:11 AM Reply Like
Full index of posts »
Latest Followers

StockTalks

More »

Latest Comments


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.