Forecasting is a natural extension of the time series modeling of Elastic ML. Since very expressive models are built behind the scenes and describe how data has behaved historically, it is therefore possible to project that information forward in time and predict how something should behave at a future time.

In this chapter, we will cover the following topics:

• Use cases for forecasting
• Theory of operation
• Single-metric forecasting
• Multiple-metric forecasting

First, however, we need to get an understanding of the caveats of predicting the future.