Re: [amibroker] The EASIEST way to use new optimizer engines, AmiBroker Email List Archive

I was already well aware of this particular algorithms abilities and short comings and you are of course entitled to your own opinion ... In any case this still doesn't address the other issues not to mention the inability to put the prowess of todays multicore processors to work ...

----- Original Message -----
From: Tomasz Janeczko
Date: Friday, June 27, 2008 3:03 pm
Subject: Re: [amibroker] The EASIEST way to use new optimizer engines
To: amibroker@xxxxxxxxxxxxxxx

> I disagree. Plateaus are found "by design", at least using CMA-
> ES algorithm that internally uses sensitivity as stopping criterion.
>
> Best regards,
> Tomasz Janeczko
> amibroker.com
> ----- Original Message -----
> From: ftonetti@xxxxxxxxxxxxx
> To: amibroker@xxxxxxxxxxxxxxx
> Sent: Friday, June 27, 2008 8:54 PM
> Subject: Re: [amibroker] The EASIEST way to use new optimizer
> engines
>
> I agree ... This is the area that intelligent optimizers shine ...
>
> However, the plateaus you refer to won't be found except by
> accident and even less likely to be found are the middle of the
> plateaus without the additional abilities to be able to be
> driven by fitness modifiers i.e. goals, constraints etc. and
> most importantly parameter sensitivity during the in sample
> optimization process as well as having more sophisticated than
> calendar based walk forward testing for out of sample.
>
> ----- Original Message -----
> From: Tomasz Janeczko
> Date: Friday, June 27, 2008 2:34 pm
> Subject: Re: [amibroker] The EASIEST way to use new optimizer
> engines To: amibroker@xxxxxxxxxxxxxxx
>
> > Hello,
> >
> > Yes these are different solvers.
> > In principle only exhaustive search is guaranteed to find
> global
> > optimum value.
> >
> > Non-exhaustive (or "intelligent") methods will find global
> or
> > local optimum.
> > The goal is of course to find global one, but if there is a
> > single sharp peak
> > out of zillions parameter combinations, non-exhaustive
> methods
> > may fail
> > to find this single peak, but taking it form trader's
> > perspecive, finding single sharp peak
> > is useless for trading because that result would be instable
> > (too fragile) and
> > not replicable in real trading. In optimization process we
> are
> > rather looking
> > for plateau regions with stable parameters and this is the
> area
> > where intelligent methods shine.
> >
> > Best regards,
> > Tomasz Janeczko
> > amibroker.com
> > ----- Original Message -----
> > From: "cstrader"
> > To:
> > Sent: Friday, June 27, 2008 8:25 PM
> > Subject: Re: [amibroker] The EASIEST way to use new
> optimizer engines
> >
> >
> > > Thanks, that helps... basically these are all different
> > variations of
> > > "solvers," right? And they don't care what they are
> modelling
> > -- they just
> > > do the math looking for the single best solution? So it's
> > just speed that
> > > matters? Or do different solvers get different solutions?
> > >
> > > And I echo all of the recent praises... take care of
> yourself
> > Tomasz!>
> > > Chuck
> > >
> > > ----- Original Message -----
> > > From: "Tomasz Janeczko"
> > > To:
> > > Sent: Friday, June 27, 2008 2:05 PM
> > > Subject: Re: [amibroker] The EASIEST way to use new
> optimizer
> > engines>
> > >
> > >> Hello,
> > >>
> > >> These are general-purpose optimization algorithms of non-
> > linear functions.
> > >> Any trading system is a function of certain number of
> > arguments. The
> > >> inputs
> > >> are parameters and quotation data , the output is your
> > optimization target
> > >> (say CAR/MDD). And you are looking for maximum of given
> function. > >>
> > >> I provided links to scientific articles describing the
> math
> > that is
> > >> underlying these algorithms
> > >> - see the README file
> > >> http://www.amibroker.com/devlog/wp-
> > content/uploads/2008/06/readme5130.html>>
> > >> Some are based on nature (animal behavior) - PSO algorithm,
> > >> or biological process - Genetic algorithms,
> > >> and some are based on mathematical concepts derived by
> humans
> > - CMA-ES.
> > >>
> > >> These algorithms are used in many different areas,
> including
> > finance.>>
> > >> Enter "PSO finance" or "CMA-ES finance" in Google and you
> > will find lots
> > >> of info.
> > >>
> > >> Best regards,
> > >> Tomasz Janeczko
> > >> amibroker.com
> > >> ----- Original Message -----
> > >> From: "cstrader"
> > >> To:
> > >> Sent: Friday, June 27, 2008 7:39 PM
> > >> Subject: Re: [amibroker] The EASIEST way to use new
> optimizer
> > engines>>
> > >>
> > >>> Yes, and more generally, what are these systems? They
> seem
> > to mimic
> > >>> evolutionary processes... how are they applied to
> > optimization? And how
> > >>> do
> > >>> we know that they work well for our particular needs --
> are
> > financial>>> markets and Darwinian evolution really the same
> > processes? Any sources,
> > >>> references appreciated.
> > >>>
> > >>> Thanks
> > >>>
> > >>>
> > >>>
> > >>> ----- Original Message -----
> > >>> From: "Thomas Ludwig"
> > >>> To:
> > >>> Sent: Friday, June 27, 2008 12:43 PM
> > >>> Subject: Re: [amibroker] The EASIEST way to use new
> > optimizer engines
> > >>>
> > >>>
> > >>>> TJ,
> > >>>>
> > >>>> the new engines work very well. But just one question:
> > >>>>
> > >>>>> (the best one which is CMA-ES).
> > >>>>
> > >>>> Are there still cases where the other engines are
> > preferable? I mean
> > >>>> there must be a reason why you included them ;-)
> > >>>>
> > >>>> Best regards,
> > >>>>
> > >>>> Thomas
> > >>>>
> > >>>> ------------------------------------
> > >>>>
> > >>>> 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
> > >>>>
> > >>>>
> > >>>>
> > >>>
> > >>>
> > >>> ------------------------------------
> > >>>
> > >>> 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
> > >>>
> > >>>
> > >>>
> > >>
> > >> ------------------------------------
> > >>
> > >> 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
> > >>
> > >>
> > >>
> > >
> > >
> > > ------------------------------------
> > >
> > > 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
> > >
> > >
> > >
> >
> > ------------------------------------
> >
> > 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
> >
> >
> >
> >
>
>