Abstract
This study undertakes a comprehensive Monte Carlo comparative simulation to evaluate the finite-sample performance of six leading panel cointegration tests Kao, Pedroni, Hadri, Hoang, Westerlund, and Johansen with the aim of identifying the most reliable and robust test across varying data environments. The simulation framework systematically varies the number of cross-sectional units (N), time dimensions (T), and cross-sectional correlation levels (ρ), while considering both the null hypothesis of no cointegration and the alternative of cointegration. The analysis reveals that in small panels (e.g., N=10, T=10), the power of most tests is limited, though Kao (0.681) and Pedroni (0.715) exhibit moderately stronger detection ability compared to Hadri (0.620) and Johansen (0.693). However, size distortions become severe in the presence of correlation. For instance, under no cointegration at (N=20, T=30, ρ=0.6), Pedroni and Hadri record rejection rates of 0.182 and 0.221, respectively, far above the nominal level, while Kao maintains a lower distortion (0.055). As panel dimensions’ increase, test power improves substantially. At (N=20, T=30, ρ=0.0, cointegration), Pedroni (0.899), Kao (0.871), Hoang (0.828), and Johansen (0.884) demonstrate strong performance. In extremely large panels (N=100, T=500), Kao, Hoang, and Westerlund consistently achieve near-perfect power (>0.95) while maintaining relatively controlled size (<0.07), underscoring their asymptotic efficiency. By contrast, Hadri persistently exhibits oversizing in correlated panels, while Johansen, though robust in large samples, remains underpowered in smaller configurations. Overall, the study identifies Kao and Westerlund as the most balanced and dependable panel cointegration tests, offering both size control and high power across a range of conditions. Pedroni and Hoang perform competitively in small samples but are vulnerable to distortions under strong correlation. These findings provide valuable guidance for empirical researchers: Kao and Westerlund should be preferred in applied work, particularly where data IJASMT E- ISSN 2489-009X , exhibit heterogeneity or potential cross-sectional dependence, while Hadri and Johansen may be applied with caution depending on the sample structure.
References
Ahmed, S., & Snardi, S. (2009). Financial integration, shocks, and the role of financial markets in developing economies. IMF Working Paper No. 09/63. International Monetary Fund. Banerjee, A., & Carrion-i-Silvestre, J. L. (2017). Testing for cointegration in large panels with structural breaks and cross-sectional dependence. Journal of Time Series Analysis, 38(3), 454–476. Wiley. Choi, I. & T.K. Chue (2007). Subsampling Hypothesis Tests for Nonstationary Panels with Applications to Exchange Rates and Stock Prices, Journal of Applied Econometrics, 22, 233–264. Coakley, J. & A.M. Fuertes (1997). New Panel Unit Root Tests of PPP, Economics Letters, 57, 17–22. Dickey, D.A. & W.A. Fuller (1979). Distribution of the Estimates for Autoregressive Time Series with a Unit Root, Journal of the American Statistical Association, 74, 427–431. Engle, R. F., & Granger, C. W. J. (1987). Co-integration and error correction: Representation, estimation, and testing. Econometrica, 55(2), 251–276. https://doi.org/10.2307/1913236 Groen, J. J. J. (1999). Long-horizon predictability of exchange rates: Is it for real? Empirical Economics, 24(3), 451–469 Kao, C. (1999). Spurious regression and residual-based tests for cointegration in panel data. Journal of Econometrics, 90(1), 1-44. Kapetanios, G.M.H. Pesaran & T. Yamagata, (2006), Panels with Nonstationary Multifactor Error Structures, CESifo Working Paper, 1(7), 8-8. Levin, A., & Lin, C.-F. (1993). Unit root tests in panel data: New results. University of California at San Diego, Discussion Paper No. 93-56. Lütkepohl, H. (2005). New introduction to multiple time series analysis. Springer-Verlag. Mehmood, H. (2019). Monte Carlo Comparison of Panel Data Cointegration tests and their Economic Application. Submitted in partial fulfillment of the requirements for the Doctor of Philosophy in Econometrics at International Institute of Islamic Economics. O’Connell, P. G. J. (1998). The overvaluation of purchasing power parity. Journal of International Economics, 44(1), 1–19. https://doi.org/10.1016/S0022-1996(97)00017- 2 Pedroni, P. (1991). Critical Values for Cointegration Tests in Heterogeneous Panels with Multiple Regressors, Oxford Bulletin of Economics and Statistics, 61, 653–670. Pedroni, P. (1995). Panel cointegration: Asymptotic and finite sample properties of pooled time series Tests with an application to the PPP hypothesis. Econometric Theory, 11(3), 595– 625. https://doi.org/10.1017/S0266466600009063 Pedroni, P. (1999). Critical Values for Cointegration Tests in Heterogeneous Panels with Multiple Regressors, Oxford Bulletin of Economics and Statistics, 61, 653–670. Pedroni, P. (2004). Panel cointegration: Asymptotic and finite sample properties of pooled time series Tests with an application to the PPP hypothesis. Econometric Theory, 20(3), 597- 625. Pesaran, M. H. (2004). General diagnostic tests for cross-sectional dependence in panels. CESifo Working Paper Series, 1229. Phillips, P. C. B., & Moon, H. R. (1999). Linear regression limit theory for nonstationary panel data. Econometrica, 67(5), 1057–1111. https://doi.org/10.1111/1468-0262.00070 Phillips, P. C. B., & Ouliaris, S. (1990). Asymptotic properties of residual-based tests for cointegration. Econometrica, 58(1), 165–193. https://doi.org/10.2307/2938339 Phillips, P. C. B., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335–346. https://doi.org/10.1093/biomet/75.2.335 IJASMT E- ISSN 2489-009X , Phillips, P.C.B. & D. Sul (2003a). Dynamic Panel Estimation and Homogeneity Testing Under Cross Section Dependence, Econometrics Journal, 6, 217-259. Pierse, R., & Snell, A. (1995). Temporal aggregation and the power of tests for a unit root. Journal of Econometrics, 65(2), 333–345. https://doi.org/10.1016/0304- 4076(94)01600-F
