ABSTRACT
This chapter studies additional features of time-series data and addresses the topics of unit roots and stationarity in time series. It explains the notions of stationarity and unit roots, presents and runs commonly used tests for the detection of unit roots. The chapter outlines data transformations to remove unit roots and obtain stationary data series, and discusses the key problems when the authors include non-stationary variables in our models. Stationarity relates to the dynamic properties of variables that the authors need to understand before estimating models and forecast. In the case of strong stationarity, the distribution of the series is unspecified but it remains the same in any sequence of values within the series. The chapter needs to test whether the rent level behaves as a random walk with a drift. It illustrates how shocks are not dissolved through time but affect the series permanently in the case of the random walk model with a drift.
