statsmodels.genmod.generalized_linear_model.PredictionResultsMean.conf_int¶

PredictionResultsMean.conf_int(method='endpoint', alpha=0.05, **kwds)[source]¶

Confidence interval for the predicted value.

This is currently only available for t and z tests.

Parameters:¶

method : {"endpoint", "delta"}¶

Method for confidence interval, “m If method is “endpoint”, then the confidence interval of the linear predictor is transformed by the prediction function. If method is “delta”, then the delta-method is used. The confidence interval in this case might reach outside the range of the prediction, for example probabilities larger than one or smaller than zero.

alpha : float, optional¶

The significance level for the confidence interval. ie., The default alpha = .05 returns a 95% confidence interval.

kwds : extra keyword arguments¶

currently ignored, only for compatibility, consistent signature

Returns:¶

ci – The array has the lower and the upper limit of the confidence interval in the columns.

Return type:¶

ndarray, (k_constraints, 2)