Internal Classes

The following summarizes classes and functions that are not intended to be directly used, but of interest only for internal use or for a developer who wants to extend on existing model classes.

Module Reference

Model and Results Classes

These are the base classes for both the estimation models and the results. They are not directly useful, but layout the structure of the subclasses and define some common methods.

Model(endog[, exog])

A (predictive) statistical model.

LikelihoodModel(endog[, exog])

Likelihood model is a subclass of Model.

GenericLikelihoodModel(endog[, exog, ...])

Allows the fitting of any likelihood function via maximum likelihood.

Results(model, params, **kwd)

Class to contain model results

LikelihoodModelResults(model, params[, ...])

Class to contain results from likelihood models

ResultMixin()

Attributes:

GenericLikelihoodModelResults(model, mlefit)

A results class for the discrete dependent variable models.

ContrastResults([t, F, sd, effect, ...])

Class for results of tests of linear restrictions on coefficients in a model.
Inheritance diagram of statsmodels.base.model, statsmodels.discrete.discrete_model, statsmodels.regression.linear_model, statsmodels.miscmodels.count
Inheritance diagram of statsmodels.regression.linear_model.GLS, statsmodels.regression.linear_model.WLS, statsmodels.regression.linear_model.OLS, statsmodels.regression.linear_model.GLSAR

Linear Model

Inheritance diagram of statsmodels.regression.linear_model

Generalized Linear Model

Inheritance diagram of statsmodels.genmod.generalized_linear_model, statsmodels.genmod.families.family, statsmodels.genmod.families.links

Discrete Model

Inheritance diagram of statsmodels.discrete.discrete_model

Robust Model

Inheritance diagram of statsmodels.robust.robust_linear_model

Vector Autoregressive Model

Inheritance diagram of statsmodels.tsa.vector_ar.var_model
Last update: Nov 14, 2024