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

Re: Averages for cme_hh_trader



PureBytes Links

Trading Reference Links

At 03:38 PM 9/25/2005, cme_hh_trader wrote:
I am using different Averages and Channels as signal generators in my
systems.

I have played with quite a few Moving Averages like Jurik, Ehlers and some
more, but I am always open to new input.

So if anyone can provide some useful info, I would really appreciate it.
You may want to play with averaging via digital filters.
As you know, a Simple Moving Average (Tradestation "@Average")
is an FIR digital filter.  So is a Weighted Moving Average
(Tradestation "@WAverage"), and so are several of the filters
in John Ehlers's books.

(FIR stands for Finite Impulse Response, meaning that the
filter output goes to zero in finite time when presented
with an impulse input.  IIR filters -- Infinite Impulse
Response -- take infinite time to go to zero when presented
with an impulse input.)

IIR digital filters are a tad less commonly used in trading
systems, with the principal example being the Exponential
Moving Average (Tradestation "@XAverage").

But there are a lot of other digital filter ideas you can
explore.  One of the most exciting is the Savitzky-Golay
filter, which is an FIR filter with interesting coefficients.

<http://documents.wolfram.com/applications/digitalimage/UsersGuide/5.3.html>

   "... The Savitzky-Golay FIR smoothing filters, also known as
   polynomial smoothing or least-squares smoothing filters, are
   generalizations of the FIR moving average filters. These are
   filters that optimally (in the sense of least squares) fit a
   set of data points to polynomials of different degrees. These
   filters preserve edges far better than a moving average filter..."

Other references

<http://research.microsoft.com/users/jckrumm/SavGol/SavGol.htm>
<http://idlastro.gsfc.nasa.gov/ftp/pro/math/poly_smooth.pro>

I myself tend to avoid FIR filters and instead gravitate
to IIR filters, because they often have a transfer function
that can realized in an analogue electronic circuit built
from resistors and capacitors.  Which makes for an enjoyable
discussion under the heading "Resistor-Capacitor Circuits
for Fun and Profit."

Mark Johnson