Home » Eng Finance » S » Seasonal adjustment

Seasonal adjustment

An adjustment made to the values in a statistical time series which is intended to allow for changes due entirely to the period of time at which the data were recorded, rather than to the underlying forces in which we may in fact be interested. For example, the number of workers unemployed will be recorded in each quarter of the year, and will usually rise sharply in winter and fall in spring. If we want to evaluate the impact that macroeconomic policies are having on the economy, we will want to remove the influence of these purely seasonal factors – lay-offs of outdoor workers because of winter weather and so on – from the data. Therefore, we can try to measure the amount of unemployment which is due to the seasonal factors and eliminate the influence of these from the data. However, it is rarely possible to estimate with perfect accuracy the influence of purely seasonal factors, and we run the risk that seasonal adjustment will distort the data by removing variations in which we are interested as well as those in which we are not. Hence ‘smoothing’ time series to eliminate seasonal variation is not an uncontroversial activity.

Reference: The Penguin Dictionary of Economics, 3rd edt.