An Application of the Functional Central Limit Theorem in the Unit Root Test in Autoregressive Models

To study the limiting distribution of the test statistics used in the unit root problems, we usually need to use the known theorem Donsker (Functional central limit theorem). In this paper, we study the limiting behavior of the unit root test statistics in the AR (1) model without and with a constant term by an indirect use of the Donsker theorem where the error terms are with noise. We also consider the case when the error terms are nonwhite noise stationary and then generalize our results to the AR (p) models. Several examples are provided to clarify the issue.
To study the limiting distribution of the test statistics used in the unit root problems, we usually need to use the known theorem Donsker (Functional central limit theorem). In this paper, we study the limiting behavior of the unit root test statistics in the AR (1) model without and with a constant term by an indirect use of the Donsker theorem where the error terms are with noise. We also consider the case when the error terms are nonwhite noise stationary and then generalize our results to the AR (p) models. Several examples are provided to clarify the issue.