Panel Unit Root Tests Under Cross Sectional Dependence
نویسندگان
چکیده
In this paper alternative approaches for testing the unit root hypothesis in panel data are considered. First, a robust version of the Dickey-Fuller t-statistic under contemporaneous correlated errors is suggested. Second, the GLS t-statistic is considered, which is based on the t-statistic of the transformed model. The asymptotic power of both tests against a sequence of local alternatives is compared. To adjust for short-run serial correlation of the errors, a pre-whitening procedure is suggested that yields a test statistic with a standard normal limiting distribution as N and T tends to infinity. The test procedure is further generalized to accommodate individual specific intercepts. From our Monte Carlo simulations it turns out that the robust OLS t-statistic performs well with respect to size and power, whereas the the GLS t-statistic may suffer from severe size distortions in small and moderate sample sizes. To improve the small sample properties of the GLS test procedure, a bootstrap version of the test is available. ∗The research for this paper was carried out within research project “Unit roots and cointegration in panel data” financed by the German Research Association (DFG). We like to thank Uwe Hassler, Adina Tarcolea, the participants of the Workshop on Nonstationary Panel Data Analysis (May 2003, University of Bonn) and the Econometric Seminar at the KU Leuven (September 2003) for their comments and suggestions.
منابع مشابه
Homogenous panel unit root tests under cross sectional dependence: Finite sample modifications and the wild bootstrap
We investigate the performance of some homogenous first and second generation panel unit root tests under alternative forms of cross sectional dependence. We formalize contemporaneous correlation through factor models, spatial autoregressive error models and combinations thereof. Our findings confirm that while the first generation test of Levin, Lin, and Chu (2002) suffers from substantial siz...
متن کاملA Simes-type Panel Unit Root Test
This paper proposes a new panel unit root test based on Simes’ [Biometrika 1986, “An Improved Bonferroni Procedure for Multiple Tests of Significance”] classical intersection test. The test is robust to general patterns of cross-sectional dependence and yet straightforward to implement, only requiring p-values of time series unit root tests of the series in the panel, and no resampling. Monte C...
متن کاملPanel Unit Root Tests under Cross-sectional Dependence: an Overview
The increasing availability of new datasets where the time-series dimension and the cross-section dimension are of the same order of magnitude asks for new techniques for the analysis of this peculiar kind of data. In the panel unit root test framework, two generations of tests have been developed: a first generation whose main limit is the assumption of cross-sectional independence across unit...
متن کاملPurchasing Power Parity and the European Single Currency: Some New Evidence
The effect of the single currency on the Purchasing Power Parity (PPP) hypothesis is examined in this study for the 15 EU countries, vis a vis the US dollar, before and after the advent of the euro. Standard as well as nonlinear unit root tests are employed on the time series dimension. Unit root tests reject PPP and the highest half-lives are observed after the introduction of the single curre...
متن کاملNonparametric Rank Tests for Non-Stationary Panels
We develop a set of nonparametric rank tests for non-stationary panels based on multivariate variance ratios which use untruncated kernels. As such, the tests do not require the choice of tuning parameters associated with bandwidth or lag length and also do not require choices with respect to numbers of common factors. The tests allow for unrestricted cross-sectional dependence and dynamic hete...
متن کاملBootstrap Unit Root Tests in Panels with Cross-Sectional Dependency
We apply bootstrap methodology to unit root tests for dependent panels with N cross-sectional units and T time series observations. More speci cally, we let each panel be driven by a general linear process which may be di erent across crosssectional units, and approximate it by a nite order autoregressive integrated process of order increasing with T . As we allow the dependency among the innov...
متن کامل