Stationarity

A stationary time series is one whose statistical properties such as mean, variance, etc. do not change over time. Most statistical forecasting methods are based on the assumption that time series is stationary. But most business and economic time series are far from stationary, but we make it approximately stationary through the use of some mathematical transformations which are relatively easier to predict.

There are different types of stationary time series such as first-order stationary, second- order stationary, trend-stationary. First-order stationary series have constant mean which means it can never change with time, but other statistical parameters can change. Second-order stationarity also called weak stationary series, have a constant mean, variance and autocovariance. Other statistical parameters can change. In trend stationarity, the mean trend is deterministic and when the trend is removed from the data it becomes a stationary stochastic process.