Economic time series have repeatable or non-repeatable fluctuation. A pattern of a time series, which repeats at regular intervals every year, same direction, and similar magnitude is defined as seasonality. The seasonal component represents intra-year fluctuations that are more or less stable year after in a time series. Possible causes of these variations are a systematic and calendar related effects and include natural factors (for instance seasonalweather patterns), administrativemeasures (for example the starting and ending dates of the school year), social/cultural/religious traditions (fixed holidays such as Christmas), the length of the months (28, 29, 30 or 31 days) or quarters (90, 91 or 92 days).
Analysts, economists, police makers use time series to make conclusions and decisions in respective area. They tray to identify important features of economic series such as short term changes, directions, turning points and consistency between other economic indicators. These points are usually in interest. Sometimes seasonal movements can make these features difficult to see and this type of analysis is not easy using raw time series data.
Deterministic, TRAMO-SEATS and ARIMA-X-12 seasonal adjustment methods are analysed in this article. 1600 time serieswere simulated for solvingwhich seasonal adjustmentmethod is precise. TRAMOSEATS and ARIMA-X-12 both perform similarly for the simulated series.
Econometric models of macroeconomic indicators of Lithuania reveal that modeling with seasonal adjusted data is more accurate.