RE: DiNapoli Osc Predictor

["Wyatt" <wparkinson@xxxxxxxxxx>]

It looks like this indicator basically takes the moving average of the high
and the MA of the low and then creates a channel from them.  What I don't
get is why the channel is asymetric (ie: overbought is 2.5 and oversold is
2.0).  I was thinking of developing something similar to this indicator when
working with using MA highs & lows in a stop and return system.

As to whether this is the actual method of the Oscillator Predictor, I have
no idea since the DiNapoli Osc. Pred. is "propietary".  Most of the
references I found to it didn't explain the method behind it.  It's creator,
Joe DiNapoli, is a systems dealer at http://fibtrader.com .  According to
him, its supposed to predict in the next bar where the overbought and
oversold levels are as these would tend to provide support and resistance.
These projected levels are supposed to tell you where to place your exit
stops or limit entries.  Here's an article where he talks about the
indicator: http://www.tfnn.com/u_article05.php

Most of the charts he shows has the indicator going to one bar in the
future.  How this is supposed to be different than moving any other OS/OB
channel (such as Bollinger Band, Keltner, etc) to the right one bar is
beyond me.

If you want to shift the indicator in any direction, you use displacement.
This is done by setting the displacement in brackets just before the plot
command.  Such as:

plot1[x](Value1, "Indicator");

Where "X" is the displacement.  Negative values shift it to the right
(future) and positive values shift it to the left (past).

-----Original Message-----
From: Dave Arnold [mailto:davearnoldus@xxxxxxxxx]
Sent: Thursday, February 05, 2004 7:10 PM
To: omega-list@xxxxxxxxxx
Subject: DiNapoli Osc Predictor


hello

I just joined the group, but I have seen Archives from
a years ago where the DiNapoli Osc Predictor was
discussed. I wanted to know if anyone out there was
willing to share any files or information on how to
create the Oscillator Predictor Bands. I have written
something myself, but not quite sure if it is correct.
it takes the Osc value and predicts the price band and
seems to be correct, but wanted to see if someone else
had something more.

Code below
{Indicator: Osciallator Predictor}
Inputs: Length(8), OB(2.5), OS(-2.0);
value1=summation(High,Length-1);
value2 = (OB*Length)+value1;
Value2= value2/(Length-1);
value3=summation(Low,Length-1);
value4 = (OS*Length)+value3;
value4= value4/(Length-1);

plot1(value2,"OB");
plot2(value4,"OS");