A0745
Title: On the influence of the choice of seasonal adjustment method on forecasting national accounts aggregates across the EU
Authors: Robert Kunst - Institute for Advanced Studies (Austria) [presenting]
Martin Ertl - Institute for Advanced Studies (Austria)
Adrian Wende - Institute for Advanced Studies (Austria)
Abstract: With quarterly national accounts aggregates, policy evaluation and forecasting often focus on seasonally adjusted time series. For the most part, two methods of seasonal adjustment are used today: Variants of the moving-average X-11 method with pre-specified filters and, alternatively, the SEATS method that is based on tentatively fitted ARIMA models. It is studied which of the two methods yields more accurate forecasts of annual targets after temporal re-aggregation, against two benchmark procedures: first, forecasting exclusively based on annual data; second, modeling quarterly year-on-year growth rates. These issues are investigated both empirically and with Monte Carlo simulations. For the simulations, data-driven time-series models are considered, both univariate and multivariate generating processes, ARIMA and VAR models among others, and various versions of seasonality generators. For the empirical investigation, the focus is on GDP and on related accounts aggregates in all EU economies.
