soundgen (version 2.7.2)

nonlinStats: Nonlinear statistics

Description

Estimates the optimal number of embedding dimensions (ed), correlation dimension D2 (d2), maximum Lyapunov exponent (ml), and the results of surrogate data testing for stochasticity (sur) using the functionality of the package nonlinearTseries. This is basically just a wrapper that puts all these functions together - convenient for frame-by-frame analysis, eg by phasegram.

Usage

nonlinStats(
  x,
  t = NULL,
  pars_ed = list(time.lag = t, max.embedding.dim = 15),
  pars_d2 = list(time.lag = t, min.embedding.dim = 2, min.radius = 0.001, max.radius =
    max(abs(x)) * 2, n.points.radius = 20, theiler.window = t * 2),
  pars_ml = list(time.lag = t, min.embedding.dim = 2, radius = 0.001, theiler.window = t
    * 2),
  pars_sur = list(FUN = nonlinearTseries::timeAsymmetry, K = 1),
  nonlinStats = c("ed", "d2", "ml", "sur")
)

Arguments

x

numeric vector such as a sound or analysis frame

t

time lag in points. Defaults to the number of steps beyond which the mutual information function reaches its minimum - see timeLag

pars_ed

a list of control parameters passed to estimateEmbeddingDim

pars_d2

a list of control parameters passed to corrDim

pars_ml

a list of control parameters passed to maxLyapunov

pars_sur

a list of control parameters passed to surrogateTest

Examples

x = sin((1:200) / 5) + rnorm(200, 0, .5)
plot(x, type = 'l')
soundgen:::nonlinStats(x)