Theory
Theory In Practice
Abstract
Forecasting into the future using historical data that has been collected at regular intervals is called time series forecasting. Different time series forecasting methods are used depending on underlying patterns in the data. In this chapter, we discuss the six components of data, focusing on the following components: level (the average component of the data), trend (long-term positive or negative movement), and random variance. Error terms can be used to evaluate how well a forecast model is performing with respect to other forecasting models and also over time.