Crossing the Median--130% per year
A Trading System That's Easy, Simple, Profitable
with Automatic Risk Control
April 3, 2009--October 30, 2009
100 Stocks--3 Portfolios--Internet Set Porfolio
Weekly--Hypothetical
See Comments below
The exit rule here is to close out a position when the ranking reaches a limit extreme, in this case, the top rank, number 1. Sell. This example has the best time/profit ratio of all.
If the density of opportunities in the sample continues, you will get 15 to 20 chances a year to execute this kind of 'crossing the median' dynamic trade. In the last five weeks alone, there have been nine opportunities. Expect more of the same, intermittently, over the months and years.
I believe that crossing-the-median is the better opportunity versus buying the top ranks and selling short the bottom ranks. It suggests accelerating momentum on its face. Arriving at extreme ranks may suggest continuing rising or falling of prices but not with the speed of the median crossers.
That is a surmise--because I have done inadequate back testing and research to prove the matter one way or the other.
But as a budding system, it is attractive enough to me personally to consider allocating some funds to it. I don't think that's good advice for you (which I do not offer in any event) because if you are going to work it as a system, you must treat it as a system. That is, take all crossings as they occur. Subjective selection of candidates is not a system. You could make up some complex filters to prescreen candidates but that defeats the mission of simplicity and ease of this system and creates another system entirely.
You may have noticed in the table on the first page a profit target of 20% per trade to recur about every 50 days, giving you 137% per year total return. This could be your third exit rule. Set an automatic fixed per-cent profit target and take each one as it occurs. If it does not occur, your exit would be the next crossing the median in the opposite direction from entry.
Caveat. The study uses 30 weeks of observations. You can't call that a 'system'. More like 300 weeks would be desirable. Until I build and run the program that captures and processes that magnitude of data, the system will have to wait to be validated. Also, it may fail validation.
The sterling current record may be the results of the recent flurry of volatility that has prevailed for about a year now and may evanesce with the snows of winter. It shows dramatically in the chart and notes of the last
Systems Tips letter.
Meantime, one may dabble in apparent passing (until proven permanent) targets of opportunity with Monday-morning odds of 50/50.
The examples
Salesforce.com
|
Monster Worldwide
|
Digital River
|
Amazon
RealNetworks
Written
11/4 to 11/8/2009
Posted
11/8/2009 12:33 p.m. ET
|
|
Back
| Subscribe
|