Hello,
I have a problem for which I would like some advice. I am doing a
small project on a chaotic double pendulum and recording a time series
of the angles of the upper and lower arms. I would like to calculate
the Lyapunov exponent (or just the largest) for each time series I
generate (different starting angles for each series).

I have read that this task is very difficult so can anybody suggest
any freeware programs for this?

Thanks for any response,

Stephan.

Hello,
I have a problem for which I would like some advice. I am doing a
small project on a chaotic double pendulum and recording a time series
of the angles of the upper and lower arms. I would like to calculate
the Lyapunov exponent (or just the largest) for each time series I
generate (different starting angles for each series).

I have read that this task is very difficult so can anybody suggest
any freeware programs for this?

On 20 Feb, 15:18, sxh...@googlemail.com wrote:
Hello,
I have a problem for which I would like some advice. I am doing a
small project on a chaotic double pendulum and recording a time series
of the angles of the upper and lower arms. I would like to calculate
the Lyapunov exponent (or just the largest) for each time series I
generate (different starting angles for each series).

I have read that this task is very difficult so can anybody suggest
any freeware programs for this?


I'd recommend Tisean which contains a program that calculates lyapunov
exponents:
http://www.mpipks-dresden.mpg.de/~tisean/Tisean_3.0.1/index.html

However, you need to be able to use the unix command line and compile
the programs yourself. If you decide on Tisean, the book "Nonlinear
Time Series Analysis" by Holger Kantz and Thomas Schreiber describes
the theory behind the programs.

Saul
TISEAN is a good choice. The physionet program too is only available as