statsmodels.discrete.count_model.ZeroInflatedPoissonResults.bic

method

ZeroInflatedPoissonResults.bic()

Bayesian information criterion. -2*llf + ln(nobs)*p where p is the number of regressors including the intercept.