statsmodels.genmod.generalized_linear_model.GLM.get_distribution¶

GLM.get_distribution(params, scale=None, exog=None, exposure=None, offset=None, var_weights=1.0, n_trials=1.0)[source]¶

Return a instance of the predictive distribution.

Parameters:¶

params : array_like¶

The model parameters.

scale : scalar¶

The scale parameter.

exog : array_like¶

The predictor variable matrix.

offset : array_like or None¶

Offset variable for predicted mean.

exposure : array_like or None¶

Log(exposure) will be added to the linear prediction.

var_weights : array_like¶

1d array of variance (analytic) weights. The default is None.

n_trials : int¶

Number of trials for the binomial distribution. The default is 1 which corresponds to a Bernoulli random variable.

Returns:¶

Instance of a scipy frozen distribution based on estimated parameters. Use the rvs method to generate random values.

Return type:¶

gen

Notes

Due to the behavior of scipy.stats.distributions objects, the returned random number generator must be called with gen.rvs(n) where n is the number of observations in the data set used to fit the model. If any other value is used for n, misleading results will be produced.