statsmodels.tools.eval_measures.aic

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

Akaike information criterion

Parameters:
llf{float, numpy:array_like}

value of the loglikelihood

nobsint

number of observations

df_modelwcint

number of parameters including constant

Returns:
aicfloat

information criterion

References

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