MathGroup Archive 1999

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Knowing a unknowed time serie

  • To: mathgroup at smc.vnet.net
  • Subject: [mg19710] Re: Knowing a unknowed time serie
  • From: slinger at signal.dra.hmg.gb (Chris Slinger)
  • Date: Sat, 11 Sep 1999 16:36:11 -0400
  • Organization: DERA Malvern
  • References: <7r4vt5$9r4@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Probably the best approach is to pose the question "can the dynamics of the time
series be described by a finite number of phase space variables ?".  An
excellent discussion of this is given in Tom Mullin's book "The Nature of Chaos"
ISBN 0 19 853990 8 Oxford University Press.  Chapter 7 "Distinguishing noise
from chaos" by Holton and May is particularly apposite and uses the above to
look at various examples, using the method of delay.  Easy enough to code into
Mathematica.

-chris

In article <7r4vt5$9r4 at smc.vnet.net>, Stevemail at teleline.es (Steve ) writes:
>I have a time series, and i want to know if this series, was created by
>a deterministic process or a full random process.  And, shure, how many
>% of each.
>
>What can i do?
>
>
>Steve Hammer
>Computer Assistant
>BARCELONA, SPAIN.
>
>Stevemail at teleline.es
>


  • Prev by Date: Manipulating a list of pairs
  • Next by Date: Re: Levenberg-Marquart code
  • Previous by thread: Re: Knowing a unknowed time serie
  • Next by thread: Mathematica can't win against Tiger Woods