You are probably noticing by this time in the book that each set of machine learning techniques has an associated set of jargon, and time series is no different.
Here is an explanation of some of this jargon that will be utilized throughout the rest of the chapter:
- Time, datetime, or timestamp: This property is the temporal element of each pairing in our time series. This could be simply a time or it could be a combination of date and time (sometimes referred to as datetime or timestamp). It might also include time zone.
- Observation, measurement, signal, or random variable: This is the property that we are trying to forecast and/or otherwise analyze as a function of time.
- Seasonality: A time series, such as the time series of air passenger data, may exhibit changes that correspond to seasons (weeks, months, years, and so on). Time series that...
