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

[amibroker] Re: Fitness Criteria that incorporates Walk Forward Result



PureBytes Links

Trading Reference Links

> If you are going to use the IS performance metrics how you suggest 
> then not only don't you believe they are worthless, you believe 
they 
> are more worthwhile then I do ...

Only under the circumstances I mentioned (when the OOS is validated) -
 just an idea - interested to hear what you thought about it.

There is no good reason not to move forward.

Thanks for your answers - appreciated.

brian. 


--- In amibroker@xxxxxxxxxxxxxxx, "Fred" <ftonetti@xxx> wrote:
>
> If you are going to use the IS performance metrics how you suggest 
> then not only don't you believe they are worthless, you believe 
they 
> are more worthwhile then I do ...
> 
> WF testing is simply multiple OOS's ... If you believe in the 
latter 
> there's no reason not to believe and utilize the former unless of 
> course you believe that once you have a system that has been 
> successfully tested OOS that you'll never move at least the end 
date 
> of your IS forward and optimize again, whether you need to or not.
> 
> --- In amibroker@xxxxxxxxxxxxxxx, "brian_z111" <brian_z111@> wrote:
> >
> > Yes, that is true.
> > 
> > What I am suggesting is that if the OOS holds up then I can use 
the 
> > combined metrics of the IS and the OOS in say, Money Management 
or 
> > other downstream analysis (assuming that the system rules remain 
> > unchanged from the IS top model to the OOS validation).
> > 
> > When you say they 'fail the steps that follow' do you mean  WF 
> testing 
> > (are you an advocate of WF testing?) or do you add something 
extra 
> > (perhaps I should read your docs?)
> > 
> > brian_z
> > 
> > 
> > --- In amibroker@xxxxxxxxxxxxxxx, "Fred" <ftonetti@> wrote:
> > >
> > > Personally I don't see in sample results as totally worthless 
nor 
> as 
> > > particularly worthwhile ... I only see them as an initial 
> yardstick 
> > > which results in a decision of whether or not additional work 
is 
> > > warranted i.e. if you can't make it work in sample then there's 
> > > probably no point in looking beyond ... If you can it doesn't 
> mean 
> > it's 
> > > tradable.  Frankly I toss more systems because they fail in the 
> steps 
> > > that follow then because I can't make the in sample results 
look 
> good.
> > >
> >
>



------------------------------------

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

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/amibroker/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/amibroker/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:amibroker-digest@xxxxxxxxxxxxxxx 
    mailto:amibroker-fullfeatured@xxxxxxxxxxxxxxx

<*> To unsubscribe from this group, send an email to:
    amibroker-unsubscribe@xxxxxxxxxxxxxxx

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