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

Re: computers



PureBytes Links

Trading Reference Links

Walter,

Do you know where can one find an algorythmn to calculate the Hurst
Exponent?

James
----- Original Message -----
From: Lionel Issen <lissen@xxxxxxx>
To: <metastock@xxxxxxxxxxxxx>
Sent: Saturday, March 18, 2000 8:22 PM
Subject: Re: computers


> Walter:
> Re last paragraph of your posting:
> What is the Hurst explicative?
> How does one tell when indicators are fading in and out of effectiveness?
>
> I am no a statistician, and I feel that your statistical references are
> suitable for those with PhDs in statistics.  No offense meant,but can you
> suggest something a bit less intimidating
> Lionel Issen
> lissen@xxxxxxxxx
> ----- Original Message -----
> From: Walter Lake <wlake@xxxxxxxxx>
> To: Metastock bulletin board <metastock@xxxxxxxxxxxxx>
> Sent: Saturday, March 18, 2000 7:48 PM
> Subject: computers
>
>
> > Thanks for your emails
> >
> > Here's an email from the guy who wrote the TOPS program.
> >
> > >Thanks for your Interest in Optimizer.
> > >
> > >The Demo in http://pages.infinit.net/xltops/  contains now the
> > >ARIMEX option .
> > >It does all the univariate Box-Jenkins ARIMA models processing
> > >i.e. Identification, Estimation and Forecasting based on your
> > >data selected.
> > >Details on the approach can be found in the Box-Jenkins book cited
> > >in reference.
> > >This site among others :
> > >http://www.itl.nist.gov/div898/handbook/pmc/section4/pmc445.htm
> > >gives a good overview of the process.
> >
> > Just keep clicking on anything in blue and save to file. A complete file
> > will emerge from an engineering point of view. Very clear and concise
> > description. ARIMA is linear as is GARCH (see Demetra).
> >
> > Re financial economics, i.e. price data, Metastock is linear analysis so
> be
> > careful. The data is unexamined as a start. One of the traders that I
know
> > uses the Hurst explicative to watch how Metastock indicators fade in and
> out
> > of effectiveness. H is computationally intensive but worth the effort.
The
> H
> > P&L is still proprietary, I don't know anyone who has it.
> >
> > Best regards
> >
> > Walter
> >
> >
>
>
>