A test of the null hypothesis that a stochastic process is a random walk (possibly with drift and/or deterministic trend) against the alternative that it is stationary, under the assumption that the random disturbances in the model are white noise. An extension which accommodates some forms of serial correlation in the disturbances is the augmented Dickey–Fuller (ADF) test. The test statistics under the null have non-standard distributions and their critical values are tabulated using computer simulations.