October 31, 2011

What do you do for an encore? You have shown that using random entries can be profitable. You have shown that using random entries can be highly profitable. You have shown that using random entries can be extremely profitable and scalable to your available capital. What do you do for an encore?

Well, you show that it was not an aberration of nature, a fluke, a well-chosen data set to make your point. So you select a totally different data set, run your script over it and show whatever the results may be. Therefore, I ran the same version of the RS script on my dataset 2.

Now, doing this should be considered as shooting oneself in the foot. There is no way using modern portfolio theory, the efficient market frontier, the efficient market hypothesis or the capital market line that you should exceed by far the most expected outcome of such an endeavor. Your system, whatever it was, should crumble. It is almost an impossibility that a trading strategy, whatever it is, could survive using a totally different and unseen data set.

If you did not implode, then this would imply that your methodology had some value, somehow, somewhere. It might not be easily expressed but one thing is for sure, there are one or more trading procedures making the whole strategy profitable.

But what can I say, you be the judge, here are the results with no modifications brought to my last script:

RS DataSet 2

RS DataSet 2

(click to enlarge) 

 And here are the charts as generated by the Wealth-Lab 4 website:

(click to enlarge) 

AAU AKAM ARUN ASYS
ATML BIDU CAM CAT
COOL ETN FFIV FIRE
GMCR HK HNL IDCC
IGTE LTXC LULU MELI
MENT MFL MGH MSN
NDSN PFCB PNRA PTI
QCOR QLTY REDF RVBD
SCSS SF SFLY SHS
SPRD SVVS TLEO TPX
UA UTEK VSEA  

( click to enlarge)  


Created on ... October 31, 2011,    © Guy R. Fleury. All rights reserved.