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

Re: [EquisMetaStock Group] Re: Ergodic Theory



PureBytes Links

Trading Reference Links

thanks for the stats knowledge.
will check your site..
 
----- Original Message -----
Sent: Saturday, February 26, 2005 8:57 AM
Subject: [EquisMetaStock Group] Re: Ergodic Theory



Hi,

Ergodicity is a property, not an indicator.  If something is ergodic,
then it means (generally) the thing is stable and converges in the
long term.  To give an example, say every day you observe the returns
on the Dow.  If it is on average say 1% and, as you add more and more
and infinitely more days, it stays at 1% and the variance of this
estimate goes to zero, then it is ergodic.  No matter what happens,
the market (soon) returns to its 1% average.  If this 1% estimate
changes as more days are added, even slightly, then the market is not
ergodic.

So I guess an ergodic indicator would eventually settle on some long
term average and never deviate from it, otherwise you know it is not
an ergodic indicator!

I think markets are not likely to be ergodic......  Ergodicity
requires something called covariance stationarity which means that
the market always behaves in a certain pattern at all times, sans
noise, dictated by what happened during the past couple of days and
the distant past dying out.  There are way too many outliers, too
much noise, things like volatility clustering and stochastic
volatility that are not ergodic.

Regards
MG Ferreira
TsaTsa EOD Programmer and trading model builder
http://tsatsaeod.ferra4models.com
http://www.ferra4models.com



--- In equismetastock@xxxxxxxxxxxxxxx, "lynn14344" <darisr@xxxx> wrote:
>
> Hi Everyone,
>
> Can someone please help me to create an indicator for ergodic.
>
> Please,  please someone assist me.
>
> Thank you very much








Yahoo! Groups Sponsor
ADVERTISEMENT
click here


Yahoo! Groups Links