Chapter 8: Forecasting Using Time Series

Predicting future trends using historical data is one of the most fascinating activities that we can do with machine learning.

Making predictions based on historical data points and time series is particularly interesting and can be very useful in different industries. Forecasting can help us in predicting the future, but also in identifying anomalies in data that don't respect the expected pattern.

In this chapter, we'll focus on time series forecasting by using the ARIMA Plus algorithm. This technique can be used to predict numerical values in different fields, such as the sales of a company, the customers in a restaurant, stock prices, and the electricity consumption of a building.

To understand how ...

Get Machine Learning with BigQuery ML now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.