sampleEntropy: Sample Entropy (also known as Kolgomorov-Sinai Entropy)

Description Usage Arguments Details Value Author(s) References Examples

Description

The Sample Entropy measures the complexity of a time series. Large values of the Sample Entropy indicate high complexity whereas that smaller values characterize more regular signals.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
  sampleEntropy(corrDim.object, do.plot = TRUE)

  getSampleEntropy(x)

  ## S3 method for class 'sampleEntropy'
 plot(x, ...)

  ## S3 method for class 'sampleEntropy'
 estimate(x,
    regression.range = NULL, do.plot = TRUE, ...)

Arguments

corrDim.object

A corrDim object from which the Sample Entropy of the time series characterized by corrDim shall be estimated.

do.plot

do.plot Logical value. If TRUE (default value), a plot of the sample entropy is shown.

...

Additional parameters.

x

A sampleEntropy object.

regression.range

Vector with 2 components denoting the range where the function will perform linear regression.

Details

The sample entropy is computed using:

hq(m,r) = log(Cq(m,r)/Cq(m+1,r)),

where m is the embedding dimension and r is the radius of the neighbourhood. When computing the correlation dimensions we use the linear regions from the correlation sums in order to do the estimates. Similarly, the sample entropy hq(m,r) should not change for both various m and r.

For each embedding dimension the sample entropy is estimated by averaging

hq(m,r) = log(Cq(m,r)/Cq(m+1,r))

over the range specified by regression range in the estimate function.

Value

A sampleEntropy object that contains a list storing the sample entropy (sample.entropy), the embedding dimensions ( embedding.dims) and radius (radius) for which the sample entropy has been computed, and the order of the sample entropy (order). The sample entropy is stored as a matrix in which each row contains the computations for a given embedding dimension and each column stores the computations for a given radius.

The getSampleEntropy returns the sample entropy function depending of the radius used for the computations.

The plot function shows the graphics for the sample entropy.

The estimate function returns a vector storing the sample entropy estimate for each embedding dimension.

Author(s)

Constantino A. Garcia

References

H. Kantz and T. Schreiber: Nonlinear Time series Analysis (Cambridge university press)

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
## Not run: 
h=henon(n.sample = 15000, n.transient = 100, a = 1.4, b = 0.3,
start = c(0.78,0.8165), do.plot = FALSE)
gen.corr.dim=corrDim(time.series=h$x,min.embedding.dim=2,max.embedding.dim=9,
                     corr.order=2, time.lag=1,min.radius=0.025,
                     max.radius=0.01,n.points.radius=20, do.plot=FALSE,
                     theiler.window=10,number.boxes=100)
se=sampleEntropy(gen.corr.dim, do.plot=FALSE)
estimate(se)
## End(Not run)

nonlinearAnalysis documentation built on May 2, 2019, 6:11 p.m.