statsmodels.genmod.generalized_linear_model.GLMResults.get_distribution#
- GLMResults.get_distribution(exog=None, exposure=None, offset=None, var_weights=1.0, n_trials=1.0)[source]#
Return an instance of the predictive distribution
- Parameters:
- exogarray_like,
optional The predictor variable matrix.
- offsetarray_like,
optional Offset variable for predicted mean.
- exposurearray_like,
optional Log(exposure) will be added to the linear prediction.
- var_weightsarray_like,
optional 1d array of variance (analytic) weights. The default is 1.
- n_trials
int,optional Number of trials for the binomial distribution. The default is 1 which corresponds to a Bernoulli random variable.
- exogarray_like,
- Returns:
genInstance of a scipy frozen distribution based on estimated parameters. Use the
rvsmethod to generate random values.
Notes
Due to the behavior of
scipy.stats.distributions objects, the returned random number generator must be called withgen.rvs(n)wherenis the number of observations in the data set used to fit the model. If any other value is used forn, misleading results will be produced.