|
PureBytes Links
Trading Reference Links
|
JMO, but it doesn't make sense to me. Unless there is a flaw in the
data, you're trying to deliberately exclude events that actually
occur in trading. Even worse, you're subjectively choosing which
events to exclude, just about the worst sin you can commit in the
system-testing community.:) The result will be a system that isn't
reflective of real-world trading.
Maybe a solution like this: Instead of eliminating from testing the
extreme trades that can actually occur, maybe have a rule to address
them on the rare occasions when they come around? Along the lines
of "If my equity in this trade increases by 50% in a week, take
profits" or "If my total portfolio equity drops by -10% in a month,
reduce the size of all positions."
FWIW.
Luck,
Sebastian
--- In amibroker@xxxxxxxxxxxxxxx, "hairy_mug" <WSCHWARZ@xxxx> wrote:
>
> Thanks Graham!
>
> I have been doing that but this elimiates candidate stocks due to
> a gap...
> What I would like to do is ignore those events, not the complete
stock.
>
> I think, for backtesting, I'll look ahead and not enter a trade if
it
> will spike and the add a switch to turn this "look-ahead" off...
>
> Does this make sense?
>
> Walt
>
> --- In amibroker@xxxxxxxxxxxxxxx, Graham <kavemanperth@xxxx> wrote:
> > You could always exclude stocks that have had large gaps during
the
> > period of your backtest. The only problem here is if the gap was
due
> > to normal trading volatility (eg sudden good or bad news) then it
> > should remain in your testing as this is real trading risk.
> > If you maintain your database to include adjustments for stock
splits
> > and consolidations then you should not have any untoward gaps.
> > If you run an exploration for gaps over the stocks you can then
check
> > first on what stocks to exclude
> > Saying that here is an example of how to exclude stocks based on
gaps
> >
> > CondGap = Sum( GapUp()*( L/Ref(H,-1) ) >= 1.3 OR GapDown()*(
> > Ref(L,-1)/H ) >= 1.3 , 500 ) ==0; // using 30% gap as limit and
over
> > 500 bars assumed as test period
> >
> > Buy = YourBuyConditions AND CondGap;
> >
> >
> > Hope this helps
> >
> >
> > On Apr 2, 2005 3:00 AM, hairy_mug <WSCHWARZ@xxxx> wrote:
> > >
> > >
> > > When I back test and optimize, there are usually one or two
stocks
> > > which throw off the profit because they had one large gain.
> > > I do not want to consider results over a set threshold.
> > >
> > > Currently I have to exclude the stocks specificly...
> > > All ideas welcome!
> > >
> > > Walt
> > >
> > >
> > > 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 other support material please check also:
> > > http://www.amibroker.com/support.html
> > >
> > >
> > > Yahoo! Groups Links
> > >
> > >
> > >
> > >
> > >
> >
> >
> > --
> > Cheers
> > Graham
> > http://e-wire.net.au/~eb_kavan/
------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/Tcy2bD/SOnJAA/cosFAA/GHeqlB/TM
--------------------------------------------------------------------~->
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 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/
<*> 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/
|