Instant Insights: Deseasonalizing Your Data

When we last talked about seasonality, we discussed how important it is to take the time of year into account. If you sell primarily Winter coats, obviously you'll see a significant uptick in your sales during the cold weather months. But what if you want a more holistic view of your sales pattern over long periods without the abnormalities of seasonality? 

Seasonal adjustment is the process of removing seasonality from a time series to present a better picture of what actually happened. There are a number of different ways to do this, most of which involve fairly complex statistical analyses (take a look at this video for a primer). The resulting data is said to be deseasonalized.

Although weekly cycles and moving holidays are not seasonal, robust seasonal adjustment methods compensate for these factors as well. By seasonally adjusting a time series, it's possible to look at any period within it and understand how it performed relative to the expected performance for that period.