Seasonality - Seasonal Adjustment

Seasonal Adjustment

Seasonal adjustment is any method for removing the seasonal component of a time series. The resulting seasonally adjusted data are used, for example, when analyzing or reporting non-seasonal trends over durations rather longer than the seasonal period. An appropriate method for seasonal adjustment is chosen on the basis of a particular view taken of the decomposition of time series into components designated with names such as "trend", "cyclic", "seasonal" and "irregular", including how these interact with each other. For example, such components might act additively or multiplicatively. Thus, if a seasonal component acts additively, the adjustment method has two stages:

• estimate the seasonal component of variation in the time series, usually in a form that has a zero mean across series;
• subtract the estimated seasonal component from the original time series, leaving the seasonally adjusted series.

One particular implementation of seasonal adjustment is provided by X-12-ARIMA.