pyhctsa.operations.correlation.periodicity_wang

pyhctsa.operations.correlation.periodicity_wang(y)

Periodicity extraction measure of Wang et al. (2007).

Implements an idea based on the periodicity extraction measure proposed in [1].

The time series is detrended using a three-knot cubic regression spline and autocorrelations are computed up to one third of the time-series length. The reported frequency corresponds to the first peak in the autocorrelation function satisfying a set of conditions.

The original paper considered a single threshold of 0.01. This implementation evaluates a range of thresholds:

  • 0

  • 0.01

  • 0.1

  • 0.2

  • \(1/\sqrt{N}\)

  • \(5/\sqrt{N}\)

  • \(10/\sqrt{N}\)

where \(N\) is the length of the time series.

References

Parameters:
y : array-like

The input time series

Returns:

Dictionary containing periodicity measures for each threshold.

Return type:

dict