fit

Fitting result containers.

Classes:

Name	Description
`FitState`	Immutable fitting result.

Classes¶

FitState¶

Immutable fitting result.

Contains all state produced by fitting a model, including coefficients, variance-covariance matrix, fitted values, residuals, and model-specific parameters.

Created by: build_fit_state(), fit_model() Consumed by: model.params, model.summary(), dispatch_infer() Augmented by: Never

Attributes:

Name	Type	Description
`coef`	`NDArray[floating]`	Coefficient estimates (1D array of length p).
`vcov`	`NDArray[floating]`	Variance-covariance matrix (p x p array).
`fitted`	`NDArray[floating]`	Fitted values (1D array of length n).
`residuals`	`NDArray[floating]`	Residuals (1D array of length n).
`leverage`	`NDArray[floating]`	Hat matrix diagonal / leverage values (1D array of length n).
`df_resid`	`float`	Residual degrees of freedom.
`loglik`	`float`	Log-likelihood at convergence.
`converged`	`bool`	Whether the optimization converged.
`n_iter`	`int`	Number of iterations (1 for closed-form solutions).
`sigma`	`float \| None`	Residual standard deviation (OLS models only).
`dispersion`	`float \| None`	Dispersion parameter (GLM models only).
`null_deviance`	`float \| None`	Null model deviance (GLM models only).
`deviance`	`float \| None`	Residual deviance, sum of unit deviances (GLM models only).
`theta`	`NDArray[floating] \| None`	Random effect variance parameters (mixed models only).
`u`	`NDArray[floating] \| None`	Spherical random effects (mixed models only).
`irls_weights`	`NDArray[floating] \| None`	IRLS weights from GLM fit (GLM sandwich estimator).
`XtWX_inv`	`NDArray[floating] \| None`	Inverse of X’WX from GLM fit (GLM sandwich estimator).

Examples:

>>> import numpy as np
>>> from fit import FitState
>>> state = FitState(
...     coef=np.array([1.0, 2.0]),
...     vcov=np.eye(2),
...     fitted=np.array([1.0, 2.0, 3.0]),
...     residuals=np.array([0.1, -0.1, 0.0]),
...     leverage=np.array([0.3, 0.3, 0.4]),
...     df_resid=1.0,
...     loglik=-10.0,
... )

Attributes¶

XtWX_inv¶

XtWX_inv: NDArray[np.floating] | None = field(default=None, validator=is_optional_ndarray)

coef¶

coef: NDArray[np.floating] = field(validator=is_ndarray)

converged¶

converged: bool = field(default=True, validator=(validators.instance_of(bool)))

deviance¶

deviance: float | None = field(default=None, validator=(validators.optional(validators.instance_of((int, float)))))

df_resid¶

df_resid: float = field(validator=(validators.instance_of((int, float))))

dispersion¶

dispersion: float | None = field(default=None, validator=(validators.optional(validators.instance_of((int, float)))))

fitted¶

fitted: NDArray[np.floating] = field(validator=is_ndarray)

irls_weights¶

irls_weights: NDArray[np.floating] | None = field(default=None, validator=is_optional_ndarray)

leverage¶

leverage: NDArray[np.floating] = field(validator=is_ndarray)

loglik¶

loglik: float = field(validator=(validators.instance_of((int, float))))

n_iter¶

n_iter: int = field(default=1, validator=(validators.instance_of(int)))

null_deviance¶

null_deviance: float | None = field(default=None, validator=(validators.optional(validators.instance_of((int, float)))))

residuals¶

residuals: NDArray[np.floating] = field(validator=is_ndarray)

sigma¶

sigma: float | None = field(default=None, validator=(validators.optional(validators.instance_of((int, float)))))

theta¶

theta: NDArray[np.floating] | None = field(default=None, validator=is_optional_ndarray)

u¶

u: NDArray[np.floating] | None = field(default=None, validator=is_optional_ndarray)

vcov¶

vcov: NDArray[np.floating] = field(validator=is_ndarray)