[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [GT] patch to correct PFEraw



Karsten Wippler wrote:
Hello Robert,

Yes, I still have access to the book, I do own it :)

The idea behind the PFE ist do measure the distance the price was moving
over time ... what it does, is more or less to calculate this distance
via pythagoras. That's where the 2 as exponent comes into the game.

You can find an explained formula at
http://www.geocities.com/user42_kevin/chart/manual/Polarized-Fractal-Efficiency.html

while http://www.geocities.com/user42_kevin/chart is worth a look
anyways.


thanks for those urls, i'll have a look see there ...

The PFE is strongly price dependent, high price -> time scale is
canceled out more or less, very low price -> price is cancelled out more
or less.

That's why the scaling is needed. So far i simply use the hint from PFE
"Set this factor to 200 for values > 1000" and divided the price by 5 to
get the correction factor. I'm still looking for a better idea.

so my PFE call is
--add=Curve(Indicators::PFE 10 5 2 {I:Generic::Divide {I:Prices Close}
5} ,[0,0,255])

thanks for this -- it is a good place to start. i see you are using
the ema smoothed variant

i think i'll add some of this discussion to the pferaw.pm and pfe.pm
pod (unless someone objects!)

aloha

ras

CU,

Karsten


<< snip >>