Bruce or Tomasz, how do i set the objective function to "objective"? I am using 5.14.
Even though the output in the AA window shows objective as a column and it computes its value, the Optimization does not seek to the objective called "objective" even when I type it into the box in AA/Backtester settings/Walk-Forward/Optimization Target. Thanks ----- Original Message ----- From: Tomasz Janeczko To: [email protected] Sent: Monday, July 07, 2008 9:00 AM Subject: Re: [amibroker] Re: Code help please... Optimize with CMAE Nice :-) Best regards, Tomasz Janeczko amibroker.com ----- Original Message ----- From: "bruce1r" <[EMAIL PROTECTED]> To: <[email protected]> Sent: Monday, July 07, 2008 5:51 PM Subject: [amibroker] Re: Code help please... Optimize with CMAE > Dennis - > > I don't post here much, but was about to post something else and saw > this. Sometimes I can't resist a challenge. It looks like you want > to experiment with CMAE as a general purpose function optimizer. > First, I concur with others - the best way is to write a DLL. But, if > you just want to play ... > > Warning, this is a hack - and hack is not being used in the good way > here. But it will allow you to play. > > 1. Setup AA to current symbol and set the Optimization target to > "objective" in AA Settings / Walk-Forward > > 2. Optimize the following code. It is set for CMAE, but you might > also find the commented out SPSO interesting. I've also done my best > to re-express your objective function to what I think you meant. > Check the comments. > _____ > > //REMEMBER - Set the Optimization target to "objective" in AA Settings > / Walk-Forward > > //OptimizerSetEngine( "spso" ); > //OptimizerSetOption( "Runs", 2 ); > //OptimizerSetOption( "MaxEval", 1000 ); > > OptimizerSetEngine("cmae"); > > X = Optimize( "X", 1, 1, 100, 0.1 ); > Y = Optimize( "Y", 1, 1, 100, 0.1 ); > > Buy = Sell = Short = Cover = 0; > > SetOption("UseCustomBacktestProc", True ); > > if ( Status( "action" ) == actionPortfolio ) > { > bo = GetBacktesterObject( ); > bo.Backtest( ); > // The original goal appeared to be to minimize the error of 100 - X*Y, > // AND to minimize the difference between X and Y. In other words, a > // result of X=10 and Y=10. > // Note that this must be expressed as an objective to be maximized. > Objective = - abs( 100 - X * Y ) - abs( X - Y ); > bo.AddCustomMetric( "objective", Objective ); > } > _____ > > It should be obvious how to setup any parameters and an objective. > Hack doesn't seem strong enough, but it was an interesting diversion. > > -- Bruce > > > --- In [email protected], "Tomasz Janeczko" <[EMAIL PROTECTED]> wrote: >> >> Hello, >> >> In that case just look at the CMAE docs (the read me is inside "cmaes") >> If this is too complicated, you may need to use SciLab for example >> www.scilab.org >> >> >> Best regards, >> Tomasz Janeczko >> amibroker.com >> ----- Original Message ----- >> From: Dennis Brown >> To: [email protected] >> Sent: Monday, July 07, 2008 4:46 PM >> Subject: Re: [amibroker] Code help please... Optimize with CMAE >> >> >> Thank you both for continuing to provide help. >> >> >> I really do not need to solve an exhaustive search for two > variables with a simplistic objective. I really need to solve > problems that have many variables and a very complex objective formula > that is not suitable for an exhaustive search. >> >> >> I only provided this simplest of all cases so the the problem > would not get in the way of my question of how to connect this simple > case to the CMAE engine. >> >> >> If that is too complicated to address on this list, then I will > have to look elsewhere for help. >> >> >> Again, thank you for your responses, >> >> >> Dennis >> >> >> On Jul 7, 2008, at 9:19 AM, Tomasz Janeczko wrote: >> >> >> Hello, >> >> Here is the simplest exhaustive search sample: >> >> function Objective( x, y ) >> { >> return sin(x) * cos(y); >> } >> >> xmin = 0; >> xmax = 100; >> xstep = 1; >> xbest = Null; >> ymin = 0; >> ymax = 100; >> ystep = 1; >> ybest = Null; >> >> >> best = -1e9; >> >> for( x = xmin; x <= xmax; x += xstep ) >> for( y = ymin; y <= ymax; y += ystep ) >> { >> f = Objective( x, y ); >> >> if( f > best ) >> { >> best = f; >> xbest = x; >> ybest = y; >> } >> } >> >> printf("Best f = %g, at x = %g, y = %g", best, xbest, ybest ); >> >> Best regards, >> Tomasz Janeczko >> amibroker.com >> ----- Original Message ----- >> From: Paul Ho >> To: [email protected] >> Sent: Monday, July 07, 2008 2:29 PM >> Subject: RE: [amibroker] Code help please... Optimize with CMAE >> >> >> I think using an exhaustive search would be a way to start. > since you only have 2 parameters, you can also put a little bit of > smart in there yourself. >> for example, if X*Y == 100 or close to that. there can be a > constraint in place to restrict the range of Y searched for a > particular value of X, and vice versa, >> Cheers >> Paul. >> >> >> >> ---------------------------------------------------------- >> From: [email protected] > [mailto:[EMAIL PROTECTED] On Behalf Of Dennis Brown >> Sent: Monday, 7 July 2008 10:17 PM >> To: [email protected] >> Subject: Re: [amibroker] Code help please... Optimize with CMAE >> >> >> Tomasz and Paul, >> >> Thank you for responding. >> >> Yes, I also had looked at those sources, but being that I am > able to >> only stumble along with C++, it is Greek to me. >> >> I think I would be better off trying to understand how to > modify the >> AmiBroker plugin DLL to call back my AFL objective function. >> >> It is still Greek, but at least it is a simpler and > meaningful Greek >> phrase to start with. >> >> I would still welcome any specific advice or hints in areas > that I am >> likely to stumble over. >> >> Best regards, >> Dennis >> >> > On Jul 7, 2008, at 7:10 AM, Tomasz Janeczko wrote: >> > >> > Yes, these sources are actually included in what you > already have on >> > your hard disk >> > under ADK\CMAE\cmaes >> > >> > Best regards, >> > Tomasz Janeczko >> > amibroker.com >> >> ----- Original Message ----- >> >> From: Paul Ho >> >> To: [email protected] >> >> Sent: Monday, July 07, 2008 12:22 PM >> >> Subject: RE: [amibroker] Code help please... Optimize > with CMAE >> >> >> >> Go to the guy's site where Tomasz download his source > code from, >> >> download his source code, and stare at that one instead, > I think >> >> its a lot closer to what you want. > http://www.bionik.tu-berlin.de/user/niko/cmaes_c.tar.gz >> >> >> >> >> >> >> On Jul 7, 2008, at 4:05 AM, Tomasz Janeczko wrote: >> >> >> >> Dennis, >> >> >> >> The optimizer plugin architecture uses backtester. You > can't go >> >> without using backtester. >> >> Call to pfEvaluateFunc invokes full-blown backtest for given >> >> parameter set. >> >> >> >> If you would like to optimize "general purpose" functions > without >> >> using backtester, >> >> you would need to take sources provided and write your > own plugin >> >> that won't >> >> use backtesting engine at all. >> >> >> >> Best regards, >> >> Tomasz Janeczko >> >> amibroker.com >> >> ----- Original Message ----- >> >> From: "Dennis Brown" <[EMAIL PROTECTED]> >> >> To: <[email protected]> >> >> Sent: Monday, July 07, 2008 3:30 AM >> >> Subject: [amibroker] Code help please... Optimize with CMAE >> >> >> >> >> >>> Hello, >> >>> >> >>> I have been staring at the CMAE DLL stuff for days and I > really need >> >>> some help to figure out how to use it in a particular way. >> >>> >> >>> I would like to use the optimizer in a generic sense to > do the >> >>> following from AFL without using the internal > backtester, meaning >> >>> only >> >>> AFL in indicator mode: >> >>> >> >>> Initialize 2 items: >> >>> item 1 is X and has a default,min,max,step,current,best > values >> >>> 1,1,1000,1,1,1 >> >>> item 2 is Y and has a default,min,max,step,current,best > values >> >>> 1,1,1000,1,1,1 >> >>> >> >>> The objective is to optimize X and Y so that X*Y==100 >> >>> >> >>> objective function in AFL: >> >>> function Objective() >> >>> { >> >>> return 100 - X*Y; >> >>> } >> >>> >> >>> The steps I would need to take as I understand them are: >> >>> >> >>> 1. Initialize the X and Y OptimizeItems by calling >> >>> OptimizerInit( with bunch of arguments) --most arguments are >> >>> irrelevant to this test. >> >>> >> >>> 2. Start the optimizer engine by calling >> >>> pfEvaluateFunc( pContext ) -- >> >>> there really is no context that I understand for this test. >> >>> >> >>> 3. The DLL calls back for the objective AFL function >> >>> >> >>> 4. It runs step 3 a number of times to find the solution > of X=Y=10 >> >>> >> >>> 5. OptimizerFinalize(same bunch of arguments as step 1) >> >>> >> >>> Of course I would prefer that step 3 is AFL calling the > optimizer >> >>> DLL >> >>> instead (simple mode), but I did not think that is how > the CMAE >> >>> works. >> >>> >> >>> Anyway, if I could get this simple case to work, I am > sure I could >> >>> figure out how do do much more complicated cases after > that on my >> >>> own. >> >>> >> >>> Of course if there is no way to use the existing DLL > without >> >>> changing >> >>> it, I would like to know that also. I should be able to > make modest >> >>> changes to the DLL myself. >> >>> >> >>> Please any hints or AFL code is appreciated. >> >>> >> >>> Best regards, >> >>> Dennis >> >>> >> > > > > ------------------------------------ > > Please note that this group is for discussion between users only. > > To get support from AmiBroker please send an e-mail directly to > SUPPORT {at} amibroker.com > > For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG: > http://www.amibroker.com/devlog/ > > For other support material please check also: > http://www.amibroker.com/support.html > Yahoo! Groups Links > > >
