Testing for unit roots: a simple alternative to Dickey—Fuller

Publisher Website
Google Scholar
Abstract
This ariticle proposes a very simple test for a unit root in a time series which is based on the residual autocorrelation function from a regression of the observed series on a linear trend. The test statistic is proportional to the maximum lag length for which the residual autocorrelations at all lower lag lengths are stickly positive. Critical values for the test are calculated using Monte Carlo simultation assuming a random walk DGP. Its robustness properties to different I(1) DGPs are examined along with its power characteristics. These are found to compare very favourably to those of Dickey–Fuller-type tests.