A Simple Integration Order Test: An Alternative to Unit Root Testing

The paper introduces order of integration test (OIT) which serves as a simple alternative to unit root built generally using auxiliary autoregressive AAR(3) model. parametric boundary conditions necessary and sufficient for testing the null hypothesis that non-stationary variable under is integrated zero I(0) were estimated via generalized least squares (GLS). decision on evaluated t-statistic. procedure was applied simulated series (y1) sample size n = 2000 known time data (y2) with two roots. results showed y1 one (I(1)) y2 I(2). These confirmed by Augmented Dickey Fuller (ADF); Phillips-Perron (PP); Kwiatkowski, Phillips, Schmidt, Shin (KPSS); Elliot, Rothenberg, Stock Point Optimal (ERS) Ng Perron (NP) tests. For logarithm transformed variable, divergent opinions other tests in clear-cut solution such makes new better alternative. Nevertheless, simplicity aptness give it leverage over conventional methods test.