statsmodels.tools.eval_measures.aicc¶

statsmodels.tools.eval_measures.aicc(llf, nobs, df_modelwc)[source]¶

Akaike information criterion (AIC) with small sample correction

Parameters:¶

llf : {float, array_like}¶

value of the loglikelihood

nobs : int¶

number of observations

df_modelwc : int¶

number of parameters including constant

Returns:¶

aicc – information criterion

Return type:¶

float

References

https://en.wikipedia.org/wiki/Akaike_information_criterion#AICc

Notes

Returns +inf if the effective degrees of freedom, defined as nobs - df_modelwc - 1.0, is <= 0.