pyhctsa.operations.entropy.approximate_entropy¶

pyhctsa.operations.entropy.approximate_entropy(x, mnom=1, rth=0.2, tau=1)¶

Approximate entropy (ApEn) of a time series.

Computes \(\mathrm{ApEn}(m, r)\), with delay vectors \((x_i, x_{i+\tau}, \ldots, x_{i+(m-1)\tau})\).

For details, see the PhysioNet documentation: https://physionet.org/physiotools/apen/

References

Parameters:¶

x : array-like¶

Input time series.

mnom : int, optional¶

Embedding dimension \(m\). Default is 1.0

rth : float, optional¶

Similarity threshold \(r\). Default is 0.2.

tau : int or str, optional¶

The time delay between the elements of a pattern: an integer, or a rule understood by get_tau() ('ac': first zero-crossing of the autocorrelation function; 'ac1e': floor of its first 1/e crossing; 'mi': the smaller of the first Kraskov automutual-information minimum and the 'ac1e' delay). The default of 1 uses consecutive samples.

Returns:¶

Approximate entropy value. NaN if the delay cannot be determined, or if the series is too short for the pattern length and delay (fewer than two delay vectors).

Return type:¶

float