statsmodels.tsa.ardl.ARDLResults.fpe# ARDLResults.fpe# Final prediction error using Lütkepohl’s definition. \(((nobs+df_{model})/(nobs-df_{model})) \sigma^2\)
