|
PureBytes Links
Trading Reference Links
|
To Monte, Paul, Neville and others that sent direct emails:
Monte, this is not the Arms' Ease of Movement Value formula
The formula I was experimenting with is simply:
Plot1: IFF (C * V <> 0, (C * V) / V,1)
Plot2: Average (Plot1,50)
Some background. I only trade Stocks: One of my charts, Price is on a
log scale and OBV runs on top of Volume bars. My hope is that the OBV 1
bar line helps me understand what is happening with Price and Volume.
But, Pull this up on SUNW weekly. You will see that while Close has
fallen from 64 and change to 22.25 yesterday, OBV has continued in a
flat range. Thus I am looking for an indicator that gives me more
sensitive price & volume information.
After entering the formula above I plotted it on top of the Volume Bars
and OBV. It was certainly more sensitive to the Price Trend, but I had
occasional spikes in low volume stocks. That is when I wrote the email
to this list.
Update:
a. As Neville pointed out: If I plot the formula in the Price subgraph,
and switch from log to linear scale, the indicator is practically the
same as the close.
b. When I plot the formula in the Volume subgraph on SUNW weekly, it
makes a strong statement that I need an indicator that is more
responsive to Price and Volume than OBV.
C. But what Price/Volume indicator to use: I stopped using Money Flow
Index because of erratic results. ex. On the same SUNW chart the 14 day
money flow index was in a steep down trend from 3/10/99 to 10/15/99,
while SUNW Close was rising before, during and after that time period.
Any suggestions would be appreciated.
And thanks for your help. Ben
"Monte C. Smith" wrote:
> I didn't read the feb 2000 and 1 issue. Is it from Arms' formula?
>
> Ben Anderson wrote:
> >
> > I am trying to create the easy language formula for the Volume-weighted
> > moving average indicator that was in the February 2001 TASC. The
> > following works, (doesn't turn the indicator off when a zero is in the
> > data field), but that zero creates a big spike down on the chart. Is
> > there a way to correct the spikes?
> >
> > IFF (C * V <> 0, (C * V) / V,1)
> >
> > Thanks for your help, Ben
|