Seeking Alpha

varan's  Instablog

varan
Send Message
Individual investor.
  • A Strategy For Trading Closed End Funds 14 comments
    Aug 28, 2013 11:59 AM

    Successful investing in Closed End Funds is somewhat of a black art, but the basket of the following CEFs together with the treasury ETFs TLT and IEF can be used for a strategy whose past performance appears to be quite attractive:

    GAB PDI PHK ETO GPM AWF BKT MMT CEF BIF MIN

    The strategy selects the top two assets every four months (first days of trading in January, May, and September) on the basis of the relative performance (total return) during the immediately preceding three months.

    The strategy yields the following results during 1991-2013:

    CAGR

    23.8%

    Sharpe Ratio

    1.17

    Sortino Ratio

    2.85

    Maximum drawdown

    12.7%

    Kelly Fraction

    .49

    It is notable that the maximum drawdown for this portfolio is considerably less than that of many balanced funds including the admirably robust VWINX (maximum drawdown 18%), one of whose objectives is to minimize the drawdowns. The returns of the strategy are, of course, substantially higher.

    The equity growth curve and the historical asset allocations are shown in the following figures.

    (click to enlarge)

    (click to enlarge)

    Disclosure: I have no positions in any stocks mentioned, but may initiate a long position in PHK, CEF over the next 72 hours.

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
  • Left Banker
    , contributor
    Comments (1677) | Send Message
     
    Nice article. Thanks.

     

    I assume you're using market return rather than NAV return, correct?

     

    Are you including distributions? Again, I assume that you are.
    28 Aug 2013, 12:28 PM Reply Like
  • varan
    , contributor
    Comments (3346) | Send Message
     
    Author’s reply » Yes, and yes. I am using adjusted close prices from the Yahoo Finance historical data, and they are based on prices adjusted for distributions.

     

    Thanks.
    28 Aug 2013, 12:30 PM Reply Like
  • Left Banker
    , contributor
    Comments (1677) | Send Message
     
    I've been thinking about this strategy since yesterday and have some thoughts/questions.

     

    CEFs tend toward low liquidity. With any substantial holdings, do you think this can be a problem at rebalancing? Seems to me with any substantial position size you're going to get beat up a bit by the spread and transaction costs at each change. I hold some CEFs. Have done well with some; not so much with ohters, I dread buying and selling them. Some have daily volumes in the 4-5 figure ranges.

     

    Do you worry at all about ex-div dates? A lot of these pay hefty distributions--it's their appeal to many investors--some of them monthly, so that ex date is always looming. On one hand you can just ignore that and assume the dividends taken or missed will even out as the price drops after the ex date.

     

    Does the total return source from yahoo you use here also include reinvestment of dividends? If not, the timing of sales/purchases with re to the ex date will matter.

     

    I understand that this is just a test of a concept, but have you ever tried to actually invest using this method?
    29 Aug 2013, 11:23 AM Reply Like
  • varan
    , contributor
    Comments (3346) | Send Message
     
    Author’s reply » Good questions. All these things will definitely degrade the actual performance, but since the CAGR is so high, the margin of safety may be sufficient.

     

    I have not tried this yet, but intend to do so starting next month.

     

    Average daily volume of most of the CEFs in this basket is over 200K. For ETO (47K), GPM (58K) and BIF (37K) it is admittedly quite low to warrant liquidity concerns especially if you commit a very large amount (which is obviously not desirable). If you remove these CEFs from the basket, the CAGR degrades by about 3%, which is not so bad.

     

    Yahoo data does include the dividends and distributions, and I am assuming that they adjust the closing prices in conformity with the ex-div dates (that is, if you had sold an asset at close on a given date, you would have received all the dividends and distributions whose ex-div dates fell before that date and after the date of purchase) as otherwise the adjustment will be meaningless.
    29 Aug 2013, 12:00 PM Reply Like
  • Left Banker
    , contributor
    Comments (1677) | Send Message
     
    I was looking into this a bit this morning and notice that PID is not a CEF. Did you mean PDI?
    2 Sep 2013, 11:49 AM Reply Like
  • varan
    , contributor
    Comments (3346) | Send Message
     
    Author’s reply » yes. my bad. it is PDI.
    2 Sep 2013, 12:17 PM Reply Like
  • Left Banker
    , contributor
    Comments (1677) | Send Message
     
    Thanks. Figured as much.
    2 Sep 2013, 12:17 PM Reply Like
  • TrendXplorer
    , contributor
    Comments (27) | Send Message
     
    Nice article. You have a great way of analyzing rotation strategies!

     

    Could you provide the stats starting January 1, 2000 and ending December 31, 2012?

     

    To get a good idea regarding the "robustness" of the strategy it would be interesting to know how the strategy did perform during these 13 years of bear-bull-bear-bull market, while there was hardly any progress for the "buy and hold" investor.
    5 Sep 2013, 05:22 AM Reply Like
  • wriskit
    , contributor
    Comments (110) | Send Message
     
    Since its all in the timing and there is no way to know whenever your arbitrary start begins what the future will be, it could just as likely turn out that you are in a bull-bear-bull-bear sequence. Would that give you the same positive results or would it be the mirror image and highly negative compared to a "buy and hold" investor who just collects dividends?

     

    I started using VWINX 10 years ago as the 60%bonds-40%equities repository for my much more volatile IRAs' RMDs. It has been a faithful companion.
    10 Oct 2013, 12:28 AM Reply Like
  • MisterJ
    , contributor
    Comments (474) | Send Message
     
    Varan,
    Thanks for your analyses on momentum rotational systems. What do you think a prolonged bond bear like 1950-80 will do to these models? I'm afraid we may not be able to rely on the bonds as going opposite stocks AND having a positive total return as in the post 1981 era. This may break a lot of the rotational system which performed so well in our recent memory. Thanks!
    20 Nov 2013, 12:09 AM Reply Like
  • varan
    , contributor
    Comments (3346) | Send Message
     
    Author’s reply » Thanks a lot.

     

    Obviously, if the negative correlations between the stocks and treasury bonds disappear, the whole set of models will fall apart. I will just have to redo all the work if some other trade-able assets that are negatively correlated with the stocks are available in case of such an event.
    20 Nov 2013, 12:25 AM Reply Like
  • MisterJ
    , contributor
    Comments (474) | Send Message
     
    I'm not so concerned about the negative correlation rather than the missing positive total return for bonds. That's where it becomes tricky- as long as the models primarily rely on the negative correlation they could still be ok, but insofar as they rely on bonds' positive total return they will not work. Can you gauge this somehow with your models? - Thanks.
    20 Nov 2013, 12:39 PM Reply Like
  • varan
    , contributor
    Comments (3346) | Send Message
     
    Author’s reply » I will take a look.

     

    Thanks for your insights, and interest.
    20 Nov 2013, 12:46 PM Reply Like
  • MisterJ
    , contributor
    Comments (474) | Send Message
     
    Thanks. This will probably be very relevant for us, and for decades to come. If you look at the post WW2 experience, the negative correlation stayed in place (stocks down in a recession and bonds up), but bonds' total return was negative. 1946-65 it was mildly negative and then until 1981 it was a rout for bonds. I would put emphasis on the 1946-65 experience in your model work. Thanks so much.
    20 Nov 2013, 06:02 PM 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.