statsmodels.gam.generalized_additive_model.GLMGamResults

class statsmodels.gam.generalized_additive_model.GLMGamResults(model, params, normalized_cov_params, scale, **kwds)

Results class for generalized additive models, GAM

This inherits from GLMResults.

All methods related to the loglikelihood function return the penalized values.

Attributes:

edfndarray

list of effective degrees of freedom for each column of the design matrix.

hat_matrix_diagndarray

diagonal of hat matrix

gcvfloat

generalized cross-validation criterion computed as gcv = scale / (1. - hat_matrix_trace / self.nobs)**2

cvfloat

cross-validation criterion computed as cv = ((resid_pearson / (1 - hat_matrix_diag))**2).sum() / nobs

Methods

conf_int([alpha])	Construct confidence interval for the fitted parameters
cov_params([r_matrix, column, scale, cov_p, ...])	Compute the variance/covariance matrix
f_test(r_matrix[, cov_p, invcov])	Compute the F-test for a joint linear hypothesis
get_distribution([exog, exposure, offset, ...])	Return an instance of the predictive distribution
get_hat_matrix_diag([observed, _axis])	Compute the diagonal of the hat matrix
get_influence([observed])	Get an instance of GLMInfluence with influence and outlier measures
get_margeff([at, method, atexog, dummy, count])	Get marginal effects of the fitted model
get_prediction([exog, exog_smooth, transform])	Compute prediction results
info_criteria(crit[, scale, dk_params])	Return an information criterion for the model
initialize(model, params, **kwargs)	Initialize (possibly re-initialize) a Results instance
llf_scaled([scale])	Return the log-likelihood at the given scale, using the estimated scale if the provided scale is None.
load(fname)	Load a pickled results instance
normalized_cov_params()	See specific model class docstring
partial_values(smooth_index[, include_constant])	Contribution of a smooth term to the linear prediction
plot_added_variable(focus_exog[, ...])	Create an added variable plot for a fitted regression model
plot_ceres_residuals(focus_exog[, frac, ...])	Conditional Expectation Partial Residuals (CERES) plot
plot_partial(smooth_index[, plot_se, cpr, ...])	Plot the contribution of a smooth term to the linear prediction
plot_partial_residuals(focus_exog[, ax])	Create a partial residual, or 'component plus residual' plot for a fitted regression model
predict([exog, exog_smooth, transform])	Compute prediction
pseudo_rsquared([kind])	Pseudo R-squared
remove_data()	Remove data arrays, all nobs arrays from result and model
save(fname[, remove_data])	Save a pickle of this instance
score_test([exog_extra, params_constrained, ...])	score test for restrictions or for omitted variables
summary([yname, xname, title, alpha])	Summarize the Regression Results
summary2([yname, xname, title, alpha, ...])	Experimental summary for regression Results
t_test(r_matrix[, cov_p, use_t])	Compute a t-test for each linear hypothesis of the form Rb = q
t_test_pairwise(term_name[, method, alpha, ...])	Perform pairwise t_test with multiple testing corrected p-values
test_significance(smooth_index)	Hypothesis test that a smooth component is zero
wald_test(r_matrix[, cov_p, invcov, use_f, ...])	Compute a Wald-test for a joint linear hypothesis
wald_test_terms([skip_single, ...])	Compute a sequence of Wald tests for terms over multiple columns

Notes

status: experimental

Warning: some inherited methods might not correctly take account of the penalization.

Methods

conf_int([alpha])	Construct confidence interval for the fitted parameters
cov_params([r_matrix, column, scale, cov_p, ...])	Compute the variance/covariance matrix
f_test(r_matrix[, cov_p, invcov])	Compute the F-test for a joint linear hypothesis
get_distribution([exog, exposure, offset, ...])	Return an instance of the predictive distribution
get_hat_matrix_diag([observed, _axis])	Compute the diagonal of the hat matrix
get_influence([observed])	Get an instance of GLMInfluence with influence and outlier measures
get_margeff([at, method, atexog, dummy, count])	Get marginal effects of the fitted model
get_prediction([exog, exog_smooth, transform])	Compute prediction results
info_criteria(crit[, scale, dk_params])	Return an information criterion for the model
initialize(model, params, **kwargs)	Initialize (possibly re-initialize) a Results instance
llf_scaled([scale])	Return the log-likelihood at the given scale, using the estimated scale if the provided scale is None.
load(fname)	Load a pickled results instance
normalized_cov_params()	See specific model class docstring
partial_values(smooth_index[, include_constant])	Contribution of a smooth term to the linear prediction
plot_added_variable(focus_exog[, ...])	Create an added variable plot for a fitted regression model
plot_ceres_residuals(focus_exog[, frac, ...])	Conditional Expectation Partial Residuals (CERES) plot
plot_partial(smooth_index[, plot_se, cpr, ...])	Plot the contribution of a smooth term to the linear prediction
plot_partial_residuals(focus_exog[, ax])	Create a partial residual, or 'component plus residual' plot for a fitted regression model
predict([exog, exog_smooth, transform])	Compute prediction
pseudo_rsquared([kind])	Pseudo R-squared
remove_data()	Remove data arrays, all nobs arrays from result and model
save(fname[, remove_data])	Save a pickle of this instance
score_test([exog_extra, params_constrained, ...])	score test for restrictions or for omitted variables
summary([yname, xname, title, alpha])	Summarize the Regression Results
summary2([yname, xname, title, alpha, ...])	Experimental summary for regression Results
t_test(r_matrix[, cov_p, use_t])	Compute a t-test for each linear hypothesis of the form Rb = q
t_test_pairwise(term_name[, method, alpha, ...])	Perform pairwise t_test with multiple testing corrected p-values
test_significance(smooth_index)	Hypothesis test that a smooth component is zero
wald_test(r_matrix[, cov_p, invcov, use_f, ...])	Compute a Wald-test for a joint linear hypothesis
wald_test_terms([skip_single, ...])	Compute a sequence of Wald tests for terms over multiple columns

Properties

aic	Akaike Information Criterion -2 * llf + 2 * (df_model + 1)
bic	Bayes Information Criterion
bic_deviance	Bayes Information Criterion
bic_llf	Bayes Information Criterion
bse	The standard errors of the parameter estimates
cv
deviance	See statsmodels.families.family for the distribution-specific deviance functions.
edf
fittedvalues	The estimated mean response.
gcv
hat_matrix_diag
hat_matrix_trace
llf	Value of the log-likelihood function evaluated at params.
llnull	Log-likelihood of the model fit with a constant as the only regressor
mu	See GLM docstring.
null	Fitted values of the null model
null_deviance	The value of the deviance function for the model fit with a constant as the only regressor
pearson_chi2	Pearson's Chi-Squared statistic is defined as the sum of the squares of the Pearson residuals.
pvalues	The two-tailed p values for the t-stats of the params
resid_anscombe	Anscombe residuals.
resid_anscombe_scaled	Scaled Anscombe residuals.
resid_anscombe_unscaled	Unscaled Anscombe residuals.
resid_deviance	Deviance residuals.
resid_pearson	Pearson residuals.
resid_response	Response residuals.
resid_working	Working residuals.
tvalues	Return the t-statistic for a given parameter estimate
use_t	Flag indicating to use the Student's distribution in inference