Statistics
Design
Contains utilities for the design of experiments.
This module defines utility functions for various experimental design methods.
- calisim.statistics.design.get_full_factorial_design(parameter_spec: ParameterSpecification) ndarray[source]
Get a full factorial design from a parameter specification.
- Parameters:
parameter_spec (ParameterSpecification) – The parameter specification.
- Returns:
The full factorial design.
- Return type:
np.ndarray
Distance Metrics
Contains utilities for calculating discrepancy values.
This module defines utility functions for discrepancies that can be used to compare simulated and observational data.
- class calisim.statistics.discrepancy.DistanceMetricBase[source]
The distance metric abstract class.
- abstract calculate(observed: ndarray, simulated: ndarray) float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) – The simulated data.
- Raises:
NotImplementedError – Error raised for the unimplemented abstract method.
- class calisim.statistics.discrepancy.GaussianLogLikelihood[source]
The Gaussian log likelihood.
- calculate(observed: ndarray, simulated: ndarray, sigma: float | ndarray | None = None) float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. sigma (float | np.ndarray | None, optional): The standard
deviation. Defaults to None.
- class calisim.statistics.discrepancy.MeanAbsoluteError[source]
The mean absolute error distance.
- calculate(observed: ndarray, simulated: ndarray, multioutput: str = 'uniform_average') float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. multioutput (str, optional): Defines aggregating of multiple output values.
Defaults to ‘uniform_average’.
- class calisim.statistics.discrepancy.MeanAbsolutePercentageError[source]
The mean absolute percentage error distance.
- calculate(observed: ndarray, simulated: ndarray, multioutput: str = 'uniform_average') float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. multioutput (str, optional): Defines aggregating of multiple output values.
Defaults to ‘uniform_average’.
- class calisim.statistics.discrepancy.MeanPinballLoss[source]
The mean pinball loss distance.
- calculate(observed: ndarray, simulated: ndarray, multioutput: str = 'uniform_average') float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. multioutput (str, optional): Defines aggregating of multiple output values.
Defaults to ‘uniform_average’.
- class calisim.statistics.discrepancy.MeanSquaredError[source]
The mean squared error distance.
- calculate(observed: ndarray, simulated: ndarray, multioutput: str = 'uniform_average') float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. multioutput (str, optional): Defines aggregating of multiple output values.
Defaults to ‘uniform_average’.
- class calisim.statistics.discrepancy.MedianAbsoluteError[source]
The median absolute error distance.
- calculate(observed: ndarray, simulated: ndarray, multioutput: str = 'uniform_average') float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. multioutput (str, optional): Defines aggregating of multiple output values.
Defaults to ‘uniform_average’.
- class calisim.statistics.discrepancy.MultivariateNormalLogLikelihood[source]
The multivariate normal log likelihood.
- calculate(observed: ndarray, simulated: ndarray, cov_matrix: ndarray | None = None, jitter: float = 1e-07) float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. cov_matrix (np.ndarray | None, optional): The covariance
matrix. Defaults to None.
- jitter (float) Jitter to stabilise the inversion of
the covariance matrix.
- class calisim.statistics.discrepancy.RootMeanSquaredError[source]
The root mean squared error distance.
- calculate(observed: ndarray, simulated: ndarray, multioutput: str = 'uniform_average') float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. multioutput (str, optional): Defines aggregating of multiple output values.
Defaults to ‘uniform_average’.
- class calisim.statistics.discrepancy.StudentsTLogLikelihood[source]
The Student’s t log likelihood.
- calculate(observed: ndarray, simulated: ndarray, nu: int = 1, sigma: float | ndarray | None = None) float | ndarray[source]
Calculate the distance between observed and simulated data.
- Parameters:
observed (np.ndarray) – The observed data.
simulated (np.ndarray) –
The simulated data. nu (int, optional): The degrees of freedom for the
t distribution. Defaults to 1.
- sigma (float | np.ndarray | None, optional): The standard
deviation. Defaults to None.