Breaks in persistence in fixed-T panel data

摘要

This paper considers an autoregressive panel data model in which the autoregressive coefficient has undergone a structural break. The object of interest is the unknown breakpoint. A least squares-based estimator is proposed that is shown to be consistent when only the number of cross-section units, N, is large and the number of time periods, T, is small, thereby enabling quick detection of the onset of a new regime. (C) 2021 The Author(s). Published by Elsevier B.V.