Removing seasonality from a time series

In economic and financial analyses, which are commonly carried out on the basis of numerous indicators, the use of data presented in a seasonally adjusted form (that is, net of seasonal fluctuations), is widely used in order to be able to grasp more clearly the short-term evolution of the phenomena considered.

Seasonality, in the dynamics of a time series, is the component that repeats itself at regular intervals every year, with variations of intensity more or less similar in the same period (month, quarter, semester, and so on) of successive years; there is different intensity during the same year. Typical examples of this are a decrease in industrial production in August following holiday closures ...

Get Hands-On Machine Learning on Google Cloud Platform now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.