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

[amibroker] Re: Help on automating optimization (genetic algorithms)



PureBytes Links

Trading Reference Links

I like what Fred suggests about working from coarse to 
fine..mathematical programming systems(what we called the products 
that IBM and others sold then) that perform optimizations on general 
functions proceed that way, except they do it with a little more 
sophistication...Back in the 60s when studying Chemical Plant 
optimizations (where your excursions and boundaries were limited by 
a physical plant process)... I first heard about Fibonacci as a 
search technique (this has probably been replaced by newer methods 
say in database searches, or Google searches).  Like a binary table 
search but more advanced because it takes fewer of steps when you 
proceed say with Fibonacci numbers to step across a range of say 0 
to 100. Of course with all the computing power now a days, we could 
do brute force optimizations without much loss in time by hitting 
every step. (;>.

We also had an saying  that one should get the chemical plant to 
yield information beside it's main objective, which were products. I 
can see the steps in optimizations as being a study where you learn 
the how the function(your system) responds as you work it from 
coarse to fine, to find the global optimum. 
Best regards
Joe   


--- In amibroker@xxxxxxxxxxxxxxx, "Fred" <fctonetti@xxxx> wrote:
> One way, not necessarily the best way to do this is to work from 
> coarse to fine ...
> 
> For example lets assume you have 4 variables that could have 
values 
> of 0 to 100 which you'd like to simultaneously optimize based on 
> something to achieve the best results yielding an integer for each 
> variable.  100^4 of course might be able to be done but you might 
> have to take a week off to have the answer.  But you might also be 
> able to get at an answer more quickly without overlooking 
potentially 
> good results by first running the optimizations as 0 to 100 by 25, 
> seeing where the best results are and then scaling down the range 
and 
> the increment for the next pass i.e. if the first run shows 50 as 
the 
> best place to be then on pass 2 you might want to set the values 
as 
> 10 to 90 by 20 etc.  If you are careful how you write the code for 
> the optimization statements this can be done in one pass from an 
> AB/AA/AFL perspective.
> 
> --- In amibroker@xxxxxxxxxxxxxxx, "Gary A. Serkhoshian" 
> <serkhoshian777@xxxx> wrote:
> > Hi all,
> >  
> > I should preface this e-mail by saying that I'm not trying to 
> create a Rube Goldberg as I can go through the process ourlined 
below 
> manually.  So, if what I'm suggesting requires a Rube Goldberg 
type 
> solution let me know, and I'll let it go.
> >  
> > Essentially what I'm doing with optimizations over multiple 
> variables is prioritizing variables on importance, and optimizing 
one 
> variable at a time leaving the others static.  I just keep 
iterating 
> through all the variables until things stabilize in terms of 
standard 
> measures.  The cocktail party term is genetic algorithms, but 
between 
> us it's necesary as optimizing over multiple variables (4+)  will 
> either take too long or worse cause AmiBroker to crash due to 
memory 
> issues.
> >  
> > Here's the punchline.  Can you all give some input on ways to 
> program this in afl so as to automate the process outlined above 
> rather than me manually iterating through as it is labor intensive.
> >  
> > Many thanks,
> > Gary
> > 
> > 
> > ---------------------------------
> > Do you Yahoo!?
> > Protect your identity with Yahoo! Mail AddressGuard


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/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/