CalibrationResult

Bases: object

A class to store fit results and statistics

Initialize the class

Parameters:
  • process (Any) –

    a stochastic process instance

  • observations (DataFrame) –

    observations data

  • delta (float, default: 1.0 ) –

    sampling interval

  • method (str, default: 'mle' ) –

    choices are 'mle', 'parametric_bootstrap', 'non_parametric_bootstrap'

  • n_boot_resamples (int, default: 1000 ) –

    number bootstrap resamples

  • n_jobs (int, default: 2 ) –

    number of parallel jobs

  • rng (Generator | int | None, default: None ) –

    bootstrap random state Generator

  • bootstrap_results (DataFrame | None, default: None ) –

    a DataFrame contained the results of bootstrap procedure

Examples:

>>> ...
>>> res = process.calibrate(data)
>>> print(res.get_summary())

method: str property

Return the method attribute

observations: pd.DataFrame property

Return the observations attribute

process property

Return the process attribute

__repr__() -> str

Override the REPL output

__str__() -> str

Override the print output

get_summary() -> dict

Collect results and fit statistics:

* LogLikelihood: the log-likelihood
* n_params: number of model parameters
* n_observations: number of records used to estimate the parameters
* AIC: Akaike information criterion
* BIC: Bayesian information criterion
* HQC: Hannan–Quinn information criterion
Returns:
  • dict –

    Information about the fit.

Examples:

>>> ...
>>> res = process.calibrate(data)
>>> res.get_summary()

show_estimated_correlation() -> go.Figure

Display the correlation obtained in the bootstrap procedure

Returns:
  • Figure –

    a dynamic graph object

Examples:

>>> ...
>>> res = process.calibrate(data)
>>> res.show_estimated_correlation()

show_parameters() -> go.Figure

Display parameters and relative errors

Returns:
  • Figure –

    a dynamic graph object

Examples:

>>> ...
>>> res = process.calibrate(data)
>>> res.show_parameters()