Inference for unit roots in dynamic panels where the time dimension is fixed

This paper derives similar, asymptotic unit root tests for first-order auto regressive panel data models, assuming that the time dimension of the panel is fixed. It is shown that the limiting distributions of the test statisti cs are normal. The assumption that the time dimension is fixed allows us to derive analytical expressions for the moments of the distributions. Simila rity with respect to the initial conditions of the data generating process is achieved by including fixed effect dummy variables in the regression mod el, while similarity with respect to fixed effects in the data generating p rocess is achieved by including a linear deterministic trend for each indiv idual unit of the panel. When fixed effects or individual trends are includ ed as regressors the least squares estimator of the autoregressive paramete r is inconsistent and thus the test statistics must be appropriately adjust ed. Monte Carlo evidence suggests that the proposed tests have empirical si ze that is very close to the nominal five percent level and substantially m ore power than the corresponding unit root tests for the single time series case. (C) 1999 Elsevier Science S.A. All rights reserved.