mystiq 650 posts msg #71016  Ignore mystiq 
1/28/2009 2:09:46 AM
Brief Description:
The Volume * Price Momentum Oscillator is a combination of price and volume indicators.
Definitions, Formulas:
As its name states, Volume * Price Momentum Oscillator (V*PMO) is both a price and volume momentum oscillator. To compute it we
First, calculate today’s V*PMO input value:
VTODAY * (CPTODAY – CPTODAY1)
where
VTODAY = today’s volume
CPTODAY = today’s closing price
CPTODAY1 = yesterday’s closing price
Then smooth the values using a 3day exponential moving average (EMA):
V*PMO = EMA(3)(VTODAY,VTODAY1,VTODAY2)
Positive Development Calculation:
For this TAI, a new positive development (NPD) occurs when V*PMO crosses over zero. That is, when V*PMO x+ 0.
This TAI is no longer positive when V*PMO crosses under zero. In other words, when V*PMO x 0.
If this TAI is still positive tomorrow it will no longer be new, but will be a Cumulative Positive Development (CPD)
“As with simple momentum indicators generally, when the nperiod exponential moving average of V*PMO is positive, momentum is bullish, so we buy, entering or initiating a long position.”
========>any coding suggestions?

chetron 2,817 posts msg #71019  Ignore chetron modified 
1/28/2009 8:45:59 AM
MAYBE...
updated with mri charting.... = )
needs to go to hundredths to work right???????????????
nope, something's not right, var6 can't be zero if var4 is below 0.....
the old, " must have 1 filtering line" requirement thing.......
still not right....
divide by a 1000???????????

wcurtis 36 posts msg #71026  Ignore wcurtis 
1/28/2009 4:22:01 PM
Many thanks to you two guys! S F has a lot of unselfish talent. This is a NICE FILTER !

springhill 74 posts msg #71030  Ignore springhill modified 
1/28/2009 5:42:11 PM
Neat indicator!
How about adding a calculation to this so that when VAR 4 goes negative it charts to the bottom of the plot and when it goes positive it charts to the top of the plot? Maybe even color red for bottom and color green for top? :) A True / False plot
This instead of plotting VAR3 and VAR4.
Or, if the to zero line were plotted one could more easily see the switch?
Possible?
The guys that provide the coding on this forum are exceptional!!!

springhill 74 posts msg #71038  Ignore springhill modified 
1/28/2009 9:42:04 PM
Gave it try on doing what I suggested. Can be improved I'm sure.

mystiq 650 posts msg #71045  Ignore mystiq 
1/29/2009 1:36:07 AM
CHETRON>thanks for coding this filter>great job>great filter

chetron 2,817 posts msg #71049  Ignore chetron 
1/29/2009 6:37:12 AM
you all are very welcome. i see that springhill has the same problem as i did. i ended up having to do a "divide by 1000" to get a better response. i have updated the filter to plot your idea.
hth
enjoy

springhill 74 posts msg #71051  Ignore springhill 
1/29/2009 10:24:13 AM
I find that if I draw only one side of the True/False that it eliminates a lot of the "noise" to look at.
change from
DRAW VAR6 ON PLOT VAR5
to
DRAW VAR5

chetron 2,817 posts msg #71058  Ignore chetron 
1/29/2009 6:34:11 PM
IT IS STILL A TAD OFF. I THINK STOCKFETCHER HAS ISSUES WITH VERY BIG NUMBERS, THAT'S WHY MY DIVIDE BY A THOUSAND WORKS.
HTH

mystiq 650 posts msg #71101  Ignore mystiq 
1/31/2009 3:48:21 AM
CHETRON> made a few adjustments:
SET{VAR1,CLOSE 1 DAY AGO}
SET{VAR2,CLOSE  VAR1}
SET{VAR3,VOLUME * VAR2}
SET{VAR4a,CEMA(VAR3,3)}
set{car4,var4a / 1000}
set{NPD,count(car4 crossed over 0 from below,1)}
set{NND,count(car4 touch 0 from below and reverse,1)}
draw car4 on plot car3
add column car4
add column NPD
add column NND
>BUT...cannot get "set{NND,count(car4 touch 0 from below and reverse,1)}" to work...looking for "car4" to touch 0(zero) and THEN reverse without crossing 0(zero)....can you code this?
