×

Testing for unit roots in panel data using a GMM approach. (English) Zbl 0911.62085

Summary: For aggregated time series unit root tests are routinely applied to choose among trend and difference stationary models. Recent work demonstrates that such tests can also be applied for testing panel data. However, it is well known that disaggregated data often exhibit a considerable amount of heterogeneity so that standard tests may perform poorly.
To account for the heterogeneity in the data we allow for individual specific deterministics, that is, we let the time trends vary across the cross section units. It is shown that standard generalized method of moments (GMM) estimators suggested for the dynamic panel data model may fail to give a valid test procedure. To overcome the difficulty, a modified GMM estimator is suggested. In a Monte Carlo study the finite sample properties of the alternative tests are compared.

MSC:

62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62F12 Asymptotic properties of parametric estimators
62E20 Asymptotic distribution theory in statistics
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Ahn, S.; Schmidt, P., Efficient estimation of models for dynamic panel data, Journal of Econometrics, 69, 5-27 (1995) · Zbl 0831.62094 · doi:10.1016/0304-4076(94)01641-C
[2] Anderson, T. W.; Hsiao, C., Formulation and estimation of dynamic models using panel data, Journal of Econometrics, 18, 47-82 (1992) · Zbl 0487.62099 · doi:10.1016/0304-4076(82)90095-1
[3] Arellano, M.; Bond, S., Some tests of specification for panel data: monte carlo evidence and an application to employment equations, Review of Economic Studies, 58, 277-297 (1991) · Zbl 0719.62116 · doi:10.2307/2297968
[4] Arellano, M.; Bover, O., Annother look at the instrumental-variable estimation of error-components models, Journal of Econometrics, 68, 29-51 (1995) · Zbl 0831.62099 · doi:10.1016/0304-4076(94)01642-D
[5] Breitung, J.; Meyer, W., Testing for unit roots in panel data: are wages on different bargaining levels cointegrated?, Applied Economics, 26, 353-361 (1994) · doi:10.1080/00036849400000081
[6] Fuller, W. A., Introduction to Statistical Time Series (1976), New York: John Wiley, New York · Zbl 0353.62050
[7] Hansen, L. P., Large sample properties of generalized method of moment estimators, Econometrics, 50, 1029-1054 (1982) · Zbl 0502.62098 · doi:10.2307/1912775
[8] Harris R, Tzavalis E (1996) Inference for unit roots in dynamic panels. Working Paper, University of Exeter · Zbl 1041.62526
[9] Im K S, Peseran, M H, Shin Y (1995) Testing for unit roots in heterogenous panels Working Paper. Department of Applied Economics, University of Cambridge
[10] Lee K H, Peseran M H, Smith R (1995) Growth and convergence: A multi-country empirical analysis of the Solow growth model. Working Paper No 9531, Department of Applied Economics, University of Cambridge
[11] Levin, A.; Lin, C. F., Unit root tests in panel data: asymptotic and finite-sample properties (1994), San Diego: University of California, San Diego
[12] Newey, W. K.; West, K. D., Hypothesis testing with efficient method of moments estimation, International Economic Review, 28, 777-787 (1987) · Zbl 0676.62029 · doi:10.2307/2526578
[13] Quah D (1992) International patterns of growth: I Persistency in Cross-Country Disparities. Discussion Paper, London School of Economics
[14] Quah, D., Exploiting cross-section variation for unit root inference in dynamic data, Economics Letters, 44, 9-19 (1994) · Zbl 0796.62107 · doi:10.1016/0165-1765(93)00302-5
[15] Zhang J (1994) Performing a unit root test in panel data—with an application to OECD convergence. Discussion Paper, Chinese University of Hong Kong
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.