statsmodels.tools.eval_measures.aicc_sigma¶
- statsmodels.tools.eval_measures.aicc_sigma(sigma2, nobs, df_modelwc, islog=False)[source]¶
Akaike information criterion (AIC) with small sample correction
- Parameters:
- Returns:
- aicc
float
information criterion
- aicc
Notes
A constant has been dropped in comparison to the loglikelihood base information criteria. These should be used to compare for comparable models.
References
https://en.wikipedia.org/wiki/Akaike_information_criterion#AICc