ABSTRACT
A time series is a series of data points, or vectors, indexed according to time. Forecasting techniques are used to determine future values of a time series. These techniques are useful in IoT since a lot of the data reported by sensors are time series. The chapter begins with a discussion on the stationarity of a time series, followed by a description of the forecasting models: the moving average or smoothing model, the moving average model MA(q), the autoregressive model AR(p), ARIMA (p, d, q), decomposition models, and the vector autoregressive model VAR(p). A set of exercises and a forecasting project is given at the end of the chapter.
