pastas.stats.signatures.colwell_contingency
===========================================

.. py:function:: pastas.stats.signatures.colwell_contingency(series: pandas.Series, bins: int = 11, freq: str = 'W', method: Literal['mean'] = 'mean', normalize: bool = True) -> tuple[float, float, float]

   
   Colwell's contingency :cite:t:`colwell_predictability_1974`

   :param series: Pandas Series with DatetimeIndex and head values.
   :type series: pandas.Series
   :param bins: number of bins to determine the states of the groundwater.
   :type bins: int
   :param freq: frequency to resample the series to.
   :type freq: str, optional
   :param method: Method to use for resampling. Only "mean" is allowed now.
   :type method: str, optional
   :param normalize: normalize the time series to values between zero and one.
   :type normalize: bool, optional

   :returns: **m** -- Colwell's contingency.
   :rtype: float

   .. rubric:: 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.















   ..
       !! processed by numpydoc !!