Eugene, your comment goes to the need to have sufficiently large backtest 
database relative to the number of adjustable parameters, so that the results 
are statistically significant. How does that relate to potential 
non-stationarity of parameters? 




________________________________
From: Eugene Kononov <[email protected]>
To: [email protected]
Sent: Sat, December 4, 2010 9:59:58 PM
Subject: Re: [JBookTrader] Dynamic Parameter Optimization


On Sat, Dec 4, 2010 at 7:05 PM, ShaggsTheStud <[email protected]> wrote:

Take for example two frogs who are trying to catch a fly.  Both frogs notice 
that the fly lands on a lilypad once an hour.  So both sit on the lily pad and 
wait.  One decides to chase the fly, but the fly moves much to fast for him.  
Eventually the fly makes it back to the lilypad, where the patient frog has 
been 
waiting patiently.
>
>Let us not forget that curve fitting to the past does not guarantee results in 
>the future!  Re-optimizing over short periods of time makes your results even 
>less statistically relevant. 
>
>
>


I agree. To make it a little more scientific, I use a "loaded coin" analogy, 
instead of the frogs analogy. How do you know that a coin is loaded? Well, you 
flip it a number of times and look at the distribution of heads and tails. But 
if the number of flips is not sufficient, even a fair coin may look like a 
loaded one. Same thing with the strategy testing. We are attempting to find a 
"loaded" strategy in the sense of positive expectancy. The number of trades is 
analogous to the number of coin flips. The larger the number of trades (and the 
number of flips), the more statistically significant is the distribution. More 
formally, the error is defined as standard deviation divided by the square root 
of the number of observations. You may notice that the inverse of this term is 
found in the Performance Index metric, which is why I like it so much for 
optimization purposes. 



-- 
You received this message because you are subscribed to the Google Groups 
"JBookTrader" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/jbooktrader?hl=en.



      

-- 
You received this message because you are subscribed to the Google Groups 
"JBookTrader" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/jbooktrader?hl=en.

Reply via email to