Panel Unit Root Tests in the Presence of Cross-Sectional Dependencies: Comparison and Implications for Modelling

Abstract

Several panel unit root tests that account for cross-section dependence using a common factor structure have been proposed in the literature recently. Pesaran's (2007) cross-sectionally augmented unit root tests are designed for cases where cross-sectional dependence is due to a single factor. The Moon and Perron (2004) tests which use defactored data are similar in spirit but can account for multiple common factors. The Bai and Ng (2004a) tests allow to determine the source of nonstationarity by testing for unit roots in the common factors and the idiosyncratic factors separately. Breitung and Das (2008) and Sul (2007) propose panel unit root tests when cross-section dependence is present possibly due to common factors, but the common factor structure is not fully exploited. This article makes four contributions: (1) it compares the testing procedures in terms of similarities and differences in the data generation process, tests, null, and alternative hypotheses considered, (2) using Monte Carlo results it compares the small sample properties of the tests in models with up to two common factors, (3) it provides an application which illustrates the use of the tests, and (4) finally, it discusses the use of the tests in modelling in general.Cross-section dependence, Factor models, Non-stationary panel data, Unit root tests, C32, C33,

Euro area inflation: aggregation bias and convergence

“…Due to the nature of subtracting the factor in Bai and Ng (2004) there are more stable sizes under cross-sectional dependency and also OLS estimation (Jang and Shin 2005). Gengenbach et al (2004) also report evidence of more favourable statistical properties for Bai and Ng (2004) than alternatives including Moon and Perron (2004) and Pesaran (2007a). Additionally Bai and Ng (2004) allows us to impose the correct number of factors based on the Bai and Ng (2002) information criteria rather than arbitrarily imposing a factor structure which might not be supported by the data.…”

“…It is recommended that a panel Bayesian information criteria is used to identify the number of factors since it is more robust to cross-sectional correlation in the idiosyncratic errors. Gengenbach et al (2004) suggest PANIC allows non-stationarity to arise in either the common or idiosyncratic component, whilst Moon and Perron (2004) and Pesaran (2007a) assume common and idiosyncratic non-stationarity under the null hypothesis. It is particularly useful in our context that PANIC determines explicitly whether the non-stationarity in a series is pervasive or variable-specific.…”

Time Series Properties of the Renewable Energy Diffusion Process: Implications for Energy Policy Design and Assessment

“…However, for the series without hydro power, the robust t-statistic exhibits some size distortion due to the small sample size. 21 As the CIPS test has better finite sample coverage than the competing panel unit root tests (Gengenbach et al 2010), we base our conclusion on Pesaran's test and consider that the panel data contain a unit root.…”

Does the Nominal Exchange Rate Regime Affect the Real Interest Parity Condition?

“…In contrast, dependencies across the panel members can lead to substantial size distortions, see Osbat (2004, 2005). The test statistics are no longer standard normal and converge to non-degenerate distributions (Gengenbach, Palm and Urbain, 2004). Note that this problem is especially relevant in the analysis presented here, since real interest rate differentials are often expressed relative to the same benchmark.…”