pastas.modelstats.Statistics.rmsn#

pastas.modelstats.Statistics.rmsn(tmin: pandas.Timestamp | str | None = None, tmax: pandas.Timestamp | str | None = None, weighted: bool = False, **kwargs) → float#

Root mean squared error of the noise.

Parameters:

  • tmin (pandas.Timestamp or str, optional) – A string or pandas.Timestamp with the start date for the period (E.g. ‘1980-01-01 00:00:00’). Strings are converted to pandas.Timestamp internally.

  • tmax (pandas.Timestamp or str, optional) – A string or pandas.Timestamp with the end date for the period (E.g. ‘2020-01-01 00:00:00’). Strings are converted to pandas.Timestamp internally.

  • weighted (bool, optional) – If weighted is True, the variances are computed using the time step between observations as weights. Default is False.

Returns:

Return a float if noisemodel is present, nan if not.

Return type:

float or nan

See also

pastas.stats.rmse