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

Re: System Tester Stops



PureBytes Links

Trading Reference Links

Glen,

I agree totally.  I reverse all my positions at the opening (for every
wheat contract I bought this morning, I also closed out a short
position...just the opposite in feeders).  I never felt that "I restricted"
my systems, yet I can understand you need for testing executions outside
MetaStocks' capabilities.  I think my "real life" (25 years as broker and
CTA) has taught me to not sweat the opening range.  It tends to work itself
out.  That's why you build "slippage" into testing.  If you find that you
are getting the "short end of the stick" on fills...change brokers.  We
should all vote with our commission dollars.  If I wasn't very satisfied
with my opening fills, I'd fold the tent and move on (I'm solicited three
times a week by FCM's).  If you're looking for bad execution, try some of
the wirehouses I used to do business with (i.e., Merrill or Paine).  

Many of us (beta testers included) have urged Equis to include the feature
you have requested.  Personally, I'd love to see the next version of
MetaStock be able to handle multiple positions (in the same contract). 
I've been using this software for 12 years and Equis still doesn't
understand that people might build a system that scales up or down while
currently in a position. 

The more flexible Equis makes it software, the more the current users will
recommend it to others.

Happy Weekend,

Steve Karnish
CCT 

----------
> From: Glen Wallace <gcwallace@xxxxxxxx>
> To: MetaStock listserver <metastock@xxxxxxxxxxxxx>
> Subject: Re: System Tester Stops
> Date: Friday, August 13, 1999 1:46 PM
> 
> Steve:
> 
> Touche.  I would point out, however, that the system test shows you
results
> only for both entering and exiting at the open (you can't enter at the
open
> of one bar and exit, say, on a stop 3 points/ATRs/SDs higher). 
Therefore,
> you have to restrict your system to match MetaStock's testing
capabilities.
> Furthermore, the system test results are for entering/exiting at the
opening
> tick, whereas in real life you execute in the opening range.  Probably
not
> material, but one is not testing like they are trading.
> 
> The message to Equis, still, is we need the flexibility to test the same
way
> we trade.
> 
> Regards
> 
> 
> ----- Original Message -----
> From: Steve Karnish <kernish@xxxxxxxxxxxx>
> To: <metastock@xxxxxxxxxxxxx>
> Sent: August 13, 1999 09:40
> Subject: Re: System Tester Stops
> 
> > Glen,
> >
> > I must be one of the "handfuls" because I test and execute at the
opening
> > for a "gang" of folks.  This morning I went long wheat and short
feeders
> > at the opening range.  I do it almost everyday...seems real to me.
> >
> > Steve Karnish
> > CCT
> >
> > ----------
> > > From: Glen Wallace <gcwallace@xxxxxxxx>
> > > To: MetaStock listserver <metastock@xxxxxxxxxxxxx>
> > > Subject: Re: System Tester Stops
> > > Date: Friday, August 13, 1999 8:10 AM
> > >
> > > Absolutely.  First, toss out exiting at the high or low because this
is
> > > not reality.  If I could do this every day, I wouldn't need MetaStock
> > > because I'd be rich.  That leaves us with exiting on the open or
close.
> > > This only reflects reality for a handful of systems, and distorts
> > > system test results to the point where they are nearly useless.
> > >
> > > What do you say Equis?
> > >
> > >
> > >
> > > > I agree that this is a feature that needs to be added asap and
would
> > > > go a long way in making a good product better.
> > > > It's a very frustrating limitation to only be able to exit at a
bar's
> > > > O,H,L, or C.  This ranks as the most important feature that needs
to
> > > > be addressed in the System Tester as far as *I* am concerned.  I'd
> > > > rather be able to run a test that accurately calculates equity from
a
> > > > more robust range of scenarios on only one security at a time,
> > > > rather than to have the ability to test multiple securities with
the
> > > > current limitations...if I had to choose.  Remarkably, not
everybody's
> > > > pain threshold is exceeded only on the O,H,L, or C.
>