statsmodels.genmod.generalized_linear_model.GLM.score_test GLM.score_test(params_constrained, k_constraints=None, exog_extra=None, observed=True) [source] score test for restrictions or for omitted variables The covariance matrix for the score is based on the Hessian, i.e. observed information matrix or optionally on the expected information matrix.. Parameters: params_constrained : array_like estimated parameter of the restricted model. This can be the parameter estimate for the current w

statsmodels.genmod.generalized_linear_model.GLM.score_obs GLM.score_obs(params, scale=None) [source] score first derivative of the loglikelihood for each observation. Parameters: params : ndarray parameter at which score is evaluated scale : None or float If scale is None, then the default scale will be calculated. Default scale is defined by self.scaletype and set in fit. If scale is not None, then it is used as a fixed scale. Returns: score_obs : ndarray, 2d The first derivative of

statsmodels.genmod.generalized_linear_model.GLM.score_factor GLM.score_factor(params, scale=None) [source] weights for score for each observation This can be considered as score residuals. Parameters: params : ndarray parameter at which Hessian is evaluated scale : None or float If scale is None, then the default scale will be calculated. Default scale is defined by self.scaletype and set in fit. If scale is not None, then it is used as a fixed scale. Returns: score_factor : ndarray_1

statsmodels.genmod.generalized_linear_model.GLM.score GLM.score(params, scale=None) [source] score, first derivative of the loglikelihood function Parameters: params : ndarray parameter at which score is evaluated scale : None or float If scale is None, then the default scale will be calculated. Default scale is defined by self.scaletype and set in fit. If scale is not None, then it is used as a fixed scale. Returns: score : ndarray_1d The first derivative of the loglikelihood functi

statsmodels.genmod.generalized_linear_model.GLM.predict GLM.predict(params, exog=None, exposure=None, offset=None, linear=False) [source] Return predicted values for a design matrix Parameters: params : array-like Parameters / coefficients of a GLM. exog : array-like, optional Design / exogenous data. Is exog is None, model exog is used. exposure : array-like, optional Exposure time values, only can be used with the log link function. See notes for details. offset : array-like, option

statsmodels.genmod.generalized_linear_model.GLM.loglike GLM.loglike(*args) [source] Loglikelihood function. Each distribution family has its own loglikelihood function. See statsmodels.families.family

statsmodels.genmod.generalized_linear_model.GLM.initialize GLM.initialize() [source] Initialize a generalized linear model.

statsmodels.genmod.generalized_linear_model.GLM.information GLM.information(params, scale=None) [source] Fisher information matrix.

statsmodels.genmod.generalized_linear_model.GLM.hessian_factor GLM.hessian_factor(params, scale=None, observed=True) [source] Weights for calculating Hessian Parameters: params : ndarray parameter at which Hessian is evaluated scale : None or float If scale is None, then the default scale will be calculated. Default scale is defined by self.scaletype and set in fit. If scale is not None, then it is used as a fixed scale. observed : bool If True, then the observed Hessian is returned. I

statsmodels.genmod.generalized_linear_model.GLM.hessian GLM.hessian(params, scale=None, observed=True) [source] Hessian, second derivative of loglikelihood function Parameters: params : ndarray parameter at which Hessian is evaluated scale : None or float If scale is None, then the default scale will be calculated. Default scale is defined by self.scaletype and set in fit. If scale is not None, then it is used as a fixed scale. observed : bool If True, then the observed Hessian is retu