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

RE: Re[2]: [amibroker] Optimizing for consistent performance



PureBytes Links

Trading Reference Links


<FONT color=#000080 
size=2>Alexander,
What an excellent 
formula.
<FONT color=#000080 
size=2> 
Unfortunately, 
the AA grid does not include percentage winners and must be calculated 
separately, which can be done.  I have been after Tomasz for a while to add 
percent winners to the AA grid.  Maybe someday soon!  (Hint, 
Hint).
<FONT color=#000080 
size=2> 
Anyway, one can 
get the percentage winners into a spreadsheet in a very simple way:  use 
Dale "Digital Dingo"'s Rx program !
 
<FONT face="Vladimir Script" color=#000080 
size=5>Rick

<FONT face=Tahoma 
size=2>-----Original Message-----From: Alexander Dimitrov 
[mailto:aldim@xxxx]Sent: Thursday, October 17, 2002 3:26 
PMTo: Steve DavisSubject: Re[2]: [amibroker] Optimizing 
for consistent performanceHello Steve,Inmy 
very humble opinion you should optimize for expectation - highexpectation 
will take care of the drawdowns and at the same time willmake the 
performance consistent.Just make the optimization as you normally do it in 
the AA. Thenexport the results to Excel. In the output columns you have 
enoughinformation to calculate the 
expectancy(Avg.Win*PercentProfitable/Avg.Loss - PercentLosing)andadd 
it to a new column. Then use the 3D analyser to see plateauxfor 
expectancy.Even better, you could take a look at the expectancyfigure, 
multiplied by the number of deals.Regards,Alexander 
DimitrovSD> Rick,SD> Hermans 3D optimizer is a blast toplay 
with. I am having great funSD> exploring new ideas. You make a good 
point, sharp peaks bad, rolling hillsSD> good. I will look for 
that.SD> I'm not sure if I'm adequately expressing my original 
point. The z-axis ofSD> the 3D plot is total system return over the 
50-year test. I am suggestingSD> perhaps there is some other system 
measurement that could be used for theSD> z-axis. I need a mathematical 
expression that evaluates the return of theSD> system but also takes 
into account the volatility of the equity curve.SD> Thanks forthe 
discussion. Putting this into words helps me clarify my ownSD> 
thoughts.SD> -SteveSD>   -----Original 
Message-----SD>   From: Rick Parsons 
[mailto:RickParsons@xxxx]SD>   Sent: Thursday, 
October 17, 2002 2:23 PMSD>   To: 
amibroker@xxxxxxxxxxxxxxxSD>   Subject: RE: [amibroker] 
Optimizing for consistent performanceSD>   
Steve,SD>   Re: >>it is designed to optimize the 
parameters to maximize the totalSD> system 
return.<<SD>   Just my humble opinion but I suspect 
you are using the 3-D for the wrongSD> purposes.  If you want 
maximum returns, you can simply optimize in AA andSD> sort by % 
return.SD>   3-D is to show you which parameters arethe 
most robust.  If you find aSD> large flat plain in the 3-D that 
means there is a wide range of parametersSD> that give consistent 
returns.  This is robustness.SD>   If you use 3-D to 
find the best return, chances are it is located on aSD> peak with sharp 
drop-offs (although this is not always the case).  The sharpSD> 
drop-offs indicate that the parameters are not 
robust.SD>   To some extent, robustness helps with 
"consistent performance".SD>   
RickSD>     -----Original 
Message-----SD>     From: Steve Davis 
[mailto:sdavis@xxxx]SD>     Sent: 
Thursday, October 17, 2002 2:07 PMSD>     To: 
amibroker@xxxxxxxxxxxxxxxSD>     Subject: RE: 
[amibroker] Optimizing for consistent 
performanceSD>     
Rick,SD>     Yes, I have tried Herman's 
optimizer -- it is designed to optimze theSD> parameters to maximize 
the total system return. That is useful, but notSD> exactly what I 
want. If I optimize the system to maximize gains, theSD> drawdowns 
become horrible. I need a system that produces steady monthlySD> 
returns.SD>     You can get at least 50 years 
of historical EOD data from Yahoo. TheSD> DOW, SP500, Utilities, and 
transports all have at least 50 years of dataSD> 
available.SD>     
-SteveSD>       -----Original 
Message-----SD>       From: Rick Parsons 
[mailto:RickParsons@xxxx]SD>       
Sent: Thursday, October 17, 2002 2:01 
PMSD>       To: 
amibroker@xxxxxxxxxxxxxxxSD>       
Subject: RE: [amibroker] Optimizing for consistent 
performanceSD>       
Steve,SD>       Not exactly sure what you 
are asking.  But have you tried Herman's 3-DSD> Optimizer for 
Excel (in download area, I believe) to find the most robustSD> 
parameters?SD>       Another 
question:  Where did you get 50 years of 
data?SD>       
Thanks,SD>       
RickSD>         -----Original 
Message-----SD>         From: 
Steve Davis 
[mailto:sdavis@xxxx]SD>         
Sent: Thursday, October 17, 2002 1:34 
PMSD>         To: 
amibroker@xxxxxxxxxxxxxxxSD>         
Subject: [amibroker] Optimizing for consistent 
performanceSD>         
AB rocks! After using AB for only a month, I have developed a systemSD> 
that produces 50% risk-adjusted-returns trading the S&P500 over the last 
50SD> years. It is a short-term trading system with an average trade 
lasting 4SD> 
days.SD>         But there 
is some spikiness in the equity curve. I am having someSD> difficulty 
tuning the system to produce steady monthly cash flow. The goalSD>is 
not to maximize profits over 50 years, but to produce consistentSD> 
profitable 
results.SD>         One 
problem is how to measure the system performance so it can beSD> 
properly tuned. I do not know many important statistics of the system. 
ForSD> example, has the system ever had a losing year? What is the 
worst-caseSD> number of consecutive losing months? I don't think 
minimizing drawdown isSD> sufficient to meet my 
goals.SD>         The 
approach I have been using is an iterative process ofSD> eye-balling 
the equity curve, changing parameters, and trying again. ThereSD> must 
be a better way. Is there a reasonable way to automate this 
task?SD>         How are 
other AB users optimizing their equity 
curve?SD>         
Cheers,SD>         
-SteveSD>         Post 
AmiQuote-related messages ONLY to: 
amiquote@xxxxxxxxxxxxxxxSD>         
(Web page: <A 
href="">http://groups.yahoo.com/group/amiquote/messages/)SD>         
Check group FAQ at:SD> <A 
href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.htmlSD>         
Your use of Yahoo! Groups is subject to the Yahoo! Terms of 
Service.SD>       Post 
AmiQuote-related messages ONLY to: 
amiquote@xxxxxxxxxxxxxxxSD>       (Web 
page: <A 
href="">http://groups.yahoo.com/group/amiquote/messages/)SD>       
Check group FAQ at:SD> <A 
href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.htmlSD>       
Your use of Yahoo! Groups is subject to the Yahoo! Terms of 
Service.SD>     Post AmiQuote-related 
messages ONLY to: amiquote@xxxxxxxxxxxxxxxSD>     
(Web page: <A 
href="">http://groups.yahoo.com/group/amiquote/messages/)SD>     
Check group FAQ at:SD> <A 
href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.htmlSD>     
Your use of Yahoo! Groups is subject to the Yahoo! Terms of 
Service.SD>         
Yahoo! Groups 
SponsorSD>               
ADVERTISEMENTSD>   Post AmiQuote-related 
messages ONLY to: amiquote@xxxxxxxxxxxxxxxSD>   (Web page: <A 
href="">http://groups.yahoo.com/group/amiquote/messages/)Post 
AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: <A 
href="">http://groups.yahoo.com/group/amiquote/messages/)Check 
group FAQ at: <A 
href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
Your use of Yahoo! Groups is subject to the <A 
href="">Yahoo! Terms of Service.