Kees,
The first formula would produce the div error. I would at
least
correct it with the divisor sum method that Roy gave
us.
Rather than a statue that would be exposed to birds and radicals
could I just have a nice portrait. That way I wouldn't be exposed to
the elements. :-)
Preston
--- In equismetastock@yahoogroups.com,
"Kees Takkenberg"
<c.a.takkenberg@...> wrote:
>
>
So I better change the one I found in Metastock ( see first mail
beneath)
for this CMO formula!
>
> Is there a spot in some country we can
use to build you a statue
Preston?
> Not joking. I very much
appreciate your enduring help.
>
> Regards,Kees.
>
>
N.B. I'll look over the ppt presentation this weekend.
> ----- Original
Message -----
> From: pumrysh
> To: equismetastock@yahoogroups.com
> Sent: Friday, February 06, 2009 12:23 AM
> Subject:
[EquisMetaStock Group] Re: About Mr.Karnish CMO3/DIA
test setup.
>
>
> Here's another CMO formula.
> It incorporates an idea
from Roy inorder to avoid the Div by zero
> error. Also a link to
another Karnish ppt presentation.
>
> Preston
>
> http://www.cedarcreektrading.com/momentum052306.ppt
>
> {Chande Momentum Oscillator}
>
Periods:=Input("Periods",1,200,3);
>
Up:=Sum((C-Ref(C,-1))*(C>Ref(C,-1)),Periods);
>
Down:=Sum((Ref(C,-1)-C)*(C<Ref(C,-1)),Periods);
>
DS:=Up+Down;{divisor sum}
>
ChandeMo:=(Up-Down)/(DS+(DS=0))*100;
> ChandeMo;{end}
>
> --- In equismetastock@yahoogroups.com,
"Kees Takkenberg"
> <c.a.takkenberg@> wrote:
>
>
> > O.K. Preston,
> >
> > Things are clear to
me now.
> >
> > Thanks again for your support , your advice
and your time.
> >
> > Regards, Kees.
> >
>
>
> > ----- Original Message -----
> > From: pumrysh
> > To: equismetastock@yahoogroups.com
> > Sent: Thursday, February 05, 2009 6:45 PM
> > Subject:
[EquisMetaStock Group] Re: About Mr.Karnish CMO3/DIA
> test
setup.
> >
> >
> > Kees,
> >
>
> Steve likes to feel very close to his indicators and often
>
nicknames
> > them. The Chande Momentum Oscillator is a built-in
indicator
and
> is
> > sometimes called the CMO. Steve
calls it CMO3 because he uses 3
> > lookback periods. You can call
the built-in function by using
> > CMO(data,periods). If you
should decide to write a custom
formula
> or
> > use
Steve's long version, once it is named you would use the
> formula
> > call like this: Fml("CMO3")
> >
> > You could
also use the built-in like this: CMO(C,3);
> >
> > Since we
know there are division errors with the long version
of
> the
> > formula, I would simply use CMO(C,3); It is the same
thing.
> >
> > Preston
> >
> > --- In equismetastock@yahoogroups.com,
"Kees Takkenberg"
> > <c.a.takkenberg@> wrote:
>
> >
> > > Peston,
> > >
> > > Thank
you for your answers Preston.
> > >
> > > So the
"Sell short" was missing the Cross-function.
> > >
> >
> I understand I can write whatever a name I want the indicator
> to
be?
> > >
> > > Am I wrong in thinking that if I do
not use the original
> indicator
> > formula in a setup I
have to use the Fml ( " ") function?
> > >
> > >
Because I don' t understand that only "CMO" works .
> > >
>
> > Regards,Kees.
> > >
> > >
> >
> ----- Original Message -----
> > > From: pumrysh
>
> > To: equismetastock@yahoogroups.com
> > > Sent: Thursday, February 05, 2009 5:59 PM
> > >
Subject: [EquisMetaStock Group] Re: About Mr.Karnish CMO3/DIA
> >
test setup.
> > >
> > >
> > >
Kees,
> > >
> > > 1. Yes...but expect some division
by zero errors with this
> formula
> > >
> > >
2. You could or you could write:
> > >
> > >
{CMO3}
> > > {Steve Karnish's CMO}
> > >
CMO(C,3);
> > >
> > > 3. Much shorter to write
CMO(C,3) and not have to worry about
> the
> > > division
error.
> > >
> > > 4. Sell short:
Cross(CMO(C,3),opt1)
> > >
> > > Preston
>
> >
> > > --- In equismetastock@yahoogroups.com,
"Kees Takkenberg"
> > > <c.a.takkenberg@>
wrote:
> > > >
> > > > It seems I have some lack
in fundamentel formulae
> > understanding.
> > > But
I'm trying to do the test myself and I don't get it right!
> > >
>
> > > > E.g. the first rule is: Buy:
Cross(-opt1,CMO(C,3))
> > > >
> > > > In
the indicatorbuilder I found this formula, wich is
> > > namend
"Chande Momentum Oscillator".
> > > >
> > > >
100*((Sum(If(C,>,Ref(C,-1),(C-Ref(C,-1)),0),14))-(Sum(If
>
(C,<,Ref
> > (C,-
> > >
1),(Ref(C,-1)-C),0),14)))
/((Sum(If(C,>,Ref(C,-1),(C-Ref(C,-
> >
1)),0),14)
> > >
+(Sum(If(C,<,Ref(C,-1),(Ref(C,-1)-C),0),14))))
>
> > >
> > > > Question:
> > > >
> > > > 1. Because of the 3 periods Mr.Karnish uses, should I
change
> > the 14
> > > into a 3?
> > >
> 2. Should I rename the "Chande Momentum Oscillator"into CMO.
> >
> > 3. By doing so, the way to write the first rule (see above)
>
> will
> > > than be : Cross(-opt1, Fml( "CMO")), because I am
referring
to
> an
> > > indicator in the
Indicatorlist?
> > > > 4.Do I have to skip the (C,3)) because I
already have
changed
> a
> > 14
> > > days
period into a 3 one? Or must I see the "CMO"as a
complete
> >
entity
> > > and set the "(C,3)" stil behind it?
> >
> >
> > > > When writing the third rule , Sell short: (
CMO(C,3),opt1)
> into
> > the
> > > systemtester
"Metastock" correct me over and over again.
> > > > Is the word
Cross missing here? Because Buy to cover is
with
> > the
> > > Crossfunction and I should say to sell short is triggered
when
> > the
> > > CMO crosses the above
triggerline!
> > > >
> > > > I would also say
that I very much appreciate Mr.Karnish
> webinar.
> > >
>
> > > > And to you all out there: I would really
appreciate your
help
> > on
> > > this.
>
> > >
> > > > Regards,
> > > >
> > > > Kees Takkenberg
> > > >
> >
>
> >
>