Testing for a Unit Root in Panels with Dynamic Factors
Preprint
- 1 September 2002
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
This paper studies testing for a unit root for large n and T panels in which the cross-sectional units are correlated. To model this cross-sectional correlation, we assume that the data is generated by an unknown number of unobservable common factors. We propose unit root tests in this environment and derive their (Gaussian) asymptotic distribution under the null hypothesis of a unit root and local alternatives. We also show that these tests have no power against the same local alternatives when it is necessary to remove deterministic components. Through Monte Carlo simulations, we provide evidence on the finite sample properties of these new tests.Keywords
All Related Versions
This publication has 32 references indexed in Scilit:
- Forecasting some low‐predictability time series using diffusion indicesJournal of Forecasting, 2003
- Monetary policy in a data-rich environmentJournal of Monetary Economics, 2003
- Inferential Theory for Factor Models of Large DimensionsEconometrica, 2003
- Nonlinear IV unit root tests in panels with cross-sectional dependencyJournal of Econometrics, 2002
- Determining the Number of Factors in Approximate Factor ModelsEconometrica, 2002
- A PANIC Attack on Unit Roots and CointegrationSSRN Electronic Journal, 2001
- A new semiparametric spatial model for panel time seriesJournal of Econometrics, 2001
- An Improved Heteroskedasticity and Autocorrelation Consistent Covariance Matrix EstimatorEconometrica, 1992
- Heteroskedasticity and Autocorrelation Consistent Covariance Matrix EstimationEconometrica, 1991
- Asymptotic Theory for Principal Component AnalysisThe Annals of Mathematical Statistics, 1963