Skip to content

My implementation of Symbolic Transfer Entropy (STE): a measure of asymmetric information flow between stochastic processes.

Notifications You must be signed in to change notification settings

tysonpond/symbolic-transfer-entropy

Repository files navigation

Symbolic Transfer Entropy

My implementation of Symbolic Transfer Entropy -- a method of estimating transfer entropy which was proposed in

Symbolic Transfer Entropy

Matthaus Staniek and Klaus Lehnertz 2008

Files

In the order of which they were used:

  1. coupled_lorenz.m
    • integrates coupled lorenz systems and stores time series data in lorenz_data/ (too big; not included in repository)
    • each saved file corresponds to one simulation (i.e. one initial condition)
      • each pair of columns corresponds to one coupling parameter
      • odd columns correspond to driver (x), even columns correspond to response (y)
  2. coupled_rossler.m
    • same as coupled_lorenz.m but with rossler systems
  3. symbolic_TE.py
    • my implementation of Symbolic Transfer Entropy (STE)
  4. TE_test.py
    • computes STE for lorenz system and rossler system
    • stores entropies in lorenz_entropies.csv and rossler_entropies.csv
  5. TE_plot.py
    • recreates figures from paper
    • resulting figures are saved in figs/

About

My implementation of Symbolic Transfer Entropy (STE): a measure of asymmetric information flow between stochastic processes.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published