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)
Adrian Wende - Institute for Advanced Studies (Austria)
Martin Ertl - Institute for Advanced Studies (Austria) [presenting]
Abstract: Empirical macroeconomic research routinely relies on seasonally adjusted data. For the most part, two methods of seasonal adjustment are used today: The moving average X-11 method (and its offsprings, such as X-12 and X-13) and the SEATS method that is based on ARIMA modeling. We study which of the two classes of methods assists in obtaining more accurate forecasts of annual targets, and in which circumstances it is better not to seasonally adjust the data at all. We investigate this question empirically and with Monte Carlo simulations, and with both data-driven ARIMA models and theoretically grounded dynamic stochastic general equilibrium models.