Re: Relative Momentum Index

["A.J. Maas" <anthmaas@xxxxxxxxx>]

Note that the build-in Equis' Momentum-custom formula is 'dividing' written as follows
    (C/Ref(C,-12))*100
whereas your Momentum is 'substracting' written
    (C-Ref(C,-12))*100
both referencing to the same earlier date and where for both too the multiplying
by 100 factor is only added, since that it is used for Y-scale plotting only and
not for/of real true value for the prime contents of the formulas.

Then based on the Equis' RSI-custom formula printed in previous mail, and on the
above Momentum-custom formula, find below the RMI (note: in the way that you have
described it and below is not the same RMI version as the MSK buid-in version). 

RMI-custom

Q:=Input("RSI Time Periods",1,1000,14);
M:=Input("Momentum Time Periods",1,1000,5);
B:=Input("Field: 1=Close, 2=Open, 3=High, 4=Low, 5=Volume",1,5,1);
Bval:=If(B=1,C,If(B=2,O,If(B=3,H,If(B=4,L,V))));
Mom:=Bval-Ref(Bval,-M);

Z:=Wilders(If(ROC(Mom,1,$)>0,
    ROC(Mom,1,$),0),LastValue(Q));

Y:=Wilders(If(ROC(Mom,1,$)<0,
    Abs(ROC(Mom,1,$)),0),LastValue(Q));

RMS:=Z/Y;
RMIcust:=100-(100/(1+RMS));
Mov(RMIcust,M,S)

Regards,
Ton Maas
ms-irb@xxxxxxxxxxxxxxxx
Dismiss the ".nospam" bit (including the dot) when replying.
Homepage  http://home.planet.nl/~anthmaas


----- Original Message ----- 
From: "Owen Davies" 
To: <metastock@xxxxxxxxxxxxx>
Sent: zaterdag 13 mei 2000 17:21
Subject: Re: Relative Momentum Index


> Ton Maas wrote:
> 
> > The closest I can find is "double smoothing the standard RSI"
> > {Relative Momentum Index}
> > Mov(Mov(RSI(14),5,S),5,S)
> > {or use 2 or 3 for last period's smoothing} 
> 
> Thanks for trying, but I don't think this is it.  In the RMI(A, B),
> the A appears to be a period variable equivalent to the
> period of the RSI.  The B is a momentum variable.  It's
> something like RSI(C - Ref(C, -B), A).  However, there is
> some smoothing going on as well, and I can't tell how that
> is done, unless it's the Wilder's smoothing described
> farther down in the original reply.
> 
> That competing program from Miami allows us to do this:
> RSI(C - C[N], K), where K is the period of the RSI and [N]
> is equivalent to the Ref time variable.  It produces an
> indicator that follows pretty much the same patterns as the
> Metastock formulation of the RMI, but is extremely jagged
> by comparison.  I don't know whether or not it incorporates
> Wilder's smoothing.
> 
> Will try rewriting the function from the ground up, including
> Wilder's smoothing and see whether it makes a difference.
> 
> Thanks again.
> 
> Owen Daviess
> 
>