[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[amibroker] Re: Correctly backtesting a system



PureBytes Links

Trading Reference Links

You can get the standard deviation easily enough by exporting the 
trade list to Excel.

--- In amibroker@xxxxxxxxxxxxxxx, "nkis22" <nkishor@xxxx> wrote:
> Anthony, 
> God stuff.
> 
> "CV = 100 X standard deviation / simple average "
> 
> can be worked only if we ge the standard deviation, which
> we don't.
> 
> However, a very rough estiamte could be obtained by
> 
> abs(largest loser-largest winnner)/6, that is
> range divided by 6.
> 
> nand
> 
> 
> 
> 
> 
> --- In amibroker@xxxxxxxxxxxxxxx, "Anthony Faragasso" 
<ajf1111@xxxx> 
> wrote:
> > To begin a Trade analysis...lets compare the results of two 
> systems. 
> > Superficially, both systems appear
> > to be the same, with identical figures for net profit, total 
number 
> of
> > trades, and average profit per trade.
> > Beneath the surface, however, lies a different story.
> >  
> >  
> >                            system A .........System B
> >  
> > Net profit............$100,000.........$100,000
> > total trades.........50.....................50
> > Average Trade...$ 2,000............$ 2,000
> > Std. Dev..............$     714...........$ 5,335
> > CV.......................35.71%............266.78%
> >  
> > Here we are measuring the volatility of the average trade.  The 
> greater the
> > volatility the less stable
> > the average.  Both systems have the same average $ 2000 profit 
per 
> trade. 
> > The trades associated with
> > system A fluctuate in a tight range around its average.  Range is 
> measured
> > here by the standard Deviation of the trades profits. 
> > Based on its Standard Deviation of $ 714, the profit range for 
> system
> > A is from $ 1,286 ( 2000 - 714 ) to $ 2,714 ( 2000 + 714 ).  
System 
> B on the
> > other hand has a standard 
> > Deviation of $ 5,335, which translates into an average trade that 
> ranges
> > between $ 7,335  and -$ 3,335.
> > These are dramatically different numbers for systems that appear 
to 
> be the
> > same.  The net result: 
> > System A is the more stable system.
> >  
> > The Systems can also be evaluated based on their COEFFICIENT OF 
> VARIATIONS
> > (CV).  
> > This statistical measure is similar to standard deviation: the 
> smaller the
> > figure, the more stable the Trades. 
> > Coefficient of Variation (CV) is calculated in a percentage 
format 
> allowing
> > for easy interpretation between systems.  
> > CV is the standard deviation of a variable ( ex. Profit per 
trade ) 
> divided
> > by the average value of the variable.
> >  
> >      CV = 100 X standard deviation / simple average
> >  
> > Look for systems with coefficient of variations of 200 % or 
less.  
> > Numbers larger than this indicate instability and should raise 
your 
> concern.
> >  
> > Anthony


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Access Your PC from Anywhere - Try GoToMyPC Free!
http://us.click.yahoo.com/XyvdrB/Uq8FAA/AG3JAA/GHeqlB/TM
---------------------------------------------------------------------~->

Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx 
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/