TSW has a diagnostic called DAA (in the SEATS, ParamEters II) DAA is the difference in absolute averages between the annual means of the raw series and of the SA series and T-C component. It measures the possible bias effect induced by the log transformation. Expressed as fractions of 1% of the series level. Is it possible to implement this diagnostic in Seasonal Adjustment Diagnostics in Demetra+? If it is possible, I think it should be work for not only when series is used log-transformation but also used in the level.