Over the course of a year, seasonal events such as changes in weather, reduced or expanded harvests, major holidays, and the opening and closing of schools can cause sharp month-to-month fluctuations in a state's economic activity, especially employment levels. The seasonal adjustment procedure takes into account past experience and mathematically smoothes out month-to-month seasonal movements, making it easier to observe the cyclical and other nonseasonal movements in the data. However, uncertainties in the seasonal adjustment process introduce additional possibilities for error.
Indicators reports seasonally adjusted data as provided by the source agency. If the source agency does not adjust the data and seasonal adjustment is appropriate, the Federal Reserve Bank of Boston seasonally adjusts the data using the Census Bureau’s X-11 ARIMA (Auto-Regressive Integrated Moving Average) method. Both unadjusted and seasonally adjusted data are available for many series in the Indicators database.
In the absence of seasonally adjusted data, comparisons of annual averages or comparisons to year-earlier periods help avoid distortion from seasonal patterns.