Seasonal adjustment means the estimation of seasonal variation and the elimination of its impact from a time series. The obtained result is a seasonally adjusted time series. The trend of a time series is obtained when both seasonal variation and irregular random variation are eliminated from a time series. Trading or working day adjusted series are in turn obtained when the factors caused by the variation in the number of trading days or weekdays is eliminated from the observation of the original time series. The Tramo/Seats method is used for the seasonal adjustment of time series at Statistics Finland. In the Tramo/Seats method, preadjustment is based on a regression model (which allows for outlying observations, public holidays and the weekday structure) and the seasonal adjustment proper on an ARIMA model constructed for the time series.
An adjustment for working days takes into account influences arising from the number of working days. This means giving consideration to lengths of months, numbers of weekdays and public holidays. Figures adjusted for working days are published for industries where variation in the number of working days has a significant impact on a time series. Compared to adjustment for trading days, adjustment for working days does not take into account the effect of different days of the week.
Trend describes the long-term development in a time series. A trend series has been adjusted for seasonal and random variations, so that the effects of e.g. weather conditions or short-term labour disputes do not show in it. By contrast, permanent changes, such as growth in demand due to changed taxation, will show in a trend. The direction indicated by the end of a trend should be interpreted with caution. The latter part of a trend indicator may change once it has been updated with data for subsequent months.