The criterion that the analysis uses to select the best model. The criterion is either Akaike's Corrected Information Criterion (AICc), Akaike's Information Criterion (AIC), or Bayesian Information Criterion (BIC).
The number of rows that the analysis uses to fit the model and generate the fits.
The number of rows that the analysis omits. For example, if the series has missing values at the end, the analysis does not use the rows with missing values.
The number of observations that occur in one season. For example, if you collect data monthly and the data have a yearly pattern, then the seasonal period is 12.
The value of λ for a Box-Cox transformation that is in the specifications for the analysis.
The largest and smallest values in the search for the λ value that stabilizes the variance of the series over time.
If the analysis searches for the optimal value of λ, then the optimal λ is the unrounded value from the search. The analysis uses the rounded value for the transformation.
If the analysis searches for the optimal value of λ, then the rounded optimal λ is the value that the Box-Cox transformation uses. If the optimal λ rounds to 0.5, then the rounded value is 0.5. Otherwise, the analysis rounds λ to the nearest integer in the search interval.
The function for the transformed series from the user-specified value of λ or from the rounded value of the optimal λ.