statsmodels.genmod.families.family.InverseGaussian

class statsmodels.genmod.families.family.InverseGaussian(link=None, check_link=True)[source]

InverseGaussian exponential family.

Parameters:
linka link instance, optional

The default link for the inverse Gaussian family is the inverse squared link. Available links are InverseSquared, Inverse, Log, and Identity. See statsmodels.genmod.families.links for more information.

check_linkbool

If True (default), then and exception is raised if the link is invalid for the family. If False, then the link is not checked.

See also

statsmodels.genmod.families.family.Family

Parent class for all links.

Link Functions

Further details on links.

Notes

The inverse Gaussian distribution is sometimes referred to in the literature as the Wald distribution.

Attributes:
InverseGaussian.linka link instance

The link function of the inverse Gaussian instance

InverseGaussian.variancevarfunc instance

variance is an instance of statsmodels.genmod.families.varfuncs.mu_cubed

Methods

variance

Methods

deviance(endog, mu[, var_weights, ...])

The deviance function evaluated at (endog, mu, var_weights, freq_weights, scale) for the distribution.

fitted(lin_pred)

Fitted values based on linear predictors lin_pred.

get_distribution(mu, scale[, var_weights])

Frozen Inverse Gaussian distribution instance for given parameters

loglike(endog, mu[, var_weights, ...])

The log-likelihood function in terms of the fitted mean response.

loglike_obs(endog, mu[, var_weights, scale])

The log-likelihood function for each observation in terms of the fitted mean response for the Inverse Gaussian distribution.

predict(mu)

Linear predictors based on given mu values.

resid_anscombe(endog, mu[, var_weights, scale])

The Anscombe residuals

resid_dev(endog, mu[, var_weights, scale])

The deviance residuals

starting_mu(y)

Starting value for mu in the IRLS algorithm.

weights(mu)

Weights for IRLS steps

Properties

link

Link function for family

links

safe_links

valid

variance


Last update: Feb 28, 2024