Chapter 4: Seasonality
One quality that sets time series apart from other datasets is that very often—but not always—the data has a certain rhythm to it. That rhythm may be yearly, possibly due to the Earth's rotation around the Sun, or daily, if rooted in the Earth's rotation around its axis. The tidal cycle follows the Moon's rotation around the Earth.
Traffic congestion follows the human activity cycle throughout the day and the five-day workweek followed by the two-day weekend; financial activity follows the quarterly business cycle. Your own body follows cycles due to your heartbeat, breathing rate, and circadian rhythm. On the very small physical and very short temporal scales, the vibration of atoms is a cause of periodicity in data. Prophet calls these cycles seasonalities.
In this chapter, you will learn about all the different types of seasonalities Prophet fits by default, how to add new ones, and how to control them. In particular, we...
