pastas.stats.signatures.colwell_components#

colwell_components(series, bins=11, freq='M', method='mean', normalize=True)[source]#

Colwell predictability, constant, and contingency [colwell_1974].

Parameters
  • series (pandas.Series) – Pandas Series with DatetimeIndex and head values.

  • bins (int) – number of bins to determine the states of the groundwater.

  • freq (str, optional) – frequency to resample the series to.

  • method (str, optional) – Method to use for resampling. Only “mean” is allowed now.

  • normalize (bool, optional) – normalize the time series to values between zero and one.

Returns

p, c, m – predictability, constancy, contingency

Return type

float, float, float

Notes

The difference between the sum of entropy for each time step and possible state of the seasonal cycle, and the overall entropy across all states and time steps, divided by the logarithm of the absolute number of possible states. Entropy according to definition in information theory, see reference for details.

References

colwell_1974

Colwell, R. K. (1974). Predictability Constancy and Contingency of periodic phenomena. Ecology, 55(5), 1148–1153.