statsmodels.tsa.ardl.UECMResults.fpe

UECMResults.fpe

Final prediction error using Lütkepohl’s definition.

\(((nobs+df_{model})/(nobs-df_{model})) \sigma^2\)