A BOOTSTRAP VIEW ON DICKEY-FULLER CONTROL CHARTS FOR AR(1) SERIES

Abstract

Dickey-Fuller control charts aim at monitoring a random walk until a given time horizon to detect stationarity as early as possible. That problem appears in many fields, especially in econometrics and the analysis of economic equilibria. To improve upon asymptotic control limits (critical values), we study the bootstrap and establish its a.s. consistency for fixed alternatives. Simulations indicate that the bootstrap control chart works very well.