Estimates the largest Lyapunov exponent from the average logarithmic divergence of initially nearby trajectories in a delay-embedded phase space (Rosenstein et al. 1993). A positive exponent indicates sensitive dependence on initial conditions (chaos / local instability).
Usage
maxLyapunovExponent(
x,
delay = NULL,
dim = NULL,
sampling_rate = 1,
mean_period = NULL,
max_steps = NULL,
fit_range = NULL
)Arguments
- x
Numeric time series.
- delay, dim
Embedding delay and dimension;
NULLestimates them viatimeDelayEmbed().- sampling_rate
Sampling rate in Hz (scales the exponent to per-second).
- mean_period
Theiler window in samples excluding temporally-close neighbours;
NULLestimates it from the mean signal period.- max_steps
Number of forward steps to track divergence;
NULLuses a default derived from the series length.- fit_range
Integer vector
c(from, to)(in steps) of the divergence curve to fit the slope;NULLuses an early near-linear window.
Value
A lyapunov_exponent object with lambda (per second), the
divergence curve, fit_range, delay and dim.