Abstract
In this paper we develop a Bayesian approach to detecting unit roots in autoregressive panel data models. Our method is based on the comparison of stationary autoregressive models with and without individual deterministic trends, to their counterpart models with a unit autoregressive root. This is done under cross-sectional dependence among the error terms of the panel units. Simulation experiments are conducted with the aim to assess the performance of the suggested inferential procedure, as well as to investigate if the Bayesian model comparison approach can distinguish unit root models from stationary autoregressive models under cross-sectional dependence. The approach is applied to real exchange rate series for a panel of the G7 countries and to a panel of US nominal interest rates data.
Similar content being viewed by others
Notes
See, e.g. Kang and Ferreira (2010). However, these Bayesian methods are not focused on drawing inferences about the presence of unit roots, which is the main focus of our paper. They are mainly interested in extracting common stochastic trends and cyclical components from nonstationary (with a unit root) autoregressive panel data models.
See Elliot and Muller (2003), for a study of the effects of the initial observations on inference about unit roots. In the panel data literature, the analysis is focused on the effects of the initial conditions on the bias of the autoregressive coefficient of dynamic panel data models (see, e.g., Arellano 2003, for a survey, and Jarocinski and Marcet 2008).
Note that random walk models are frequently used both in single time series and panel data literature to describe the dynamics of many asset prices (see, e.g., Campbell et al. 1997, for a survey).
The assumption of a common autoregressive coefficient under stationarity corresponds to a classical unit root test that assumes a homogeneous autoregressive coefficient under the alternative hypothesis. This is a convenient assumption which enables us to compute analytically the marginal likelihoods of several competing models and develop a Bayesian approach to distinguishing unit root from stationary processes based on model comparison. Relaxing this assumption may increase the posterior probability of stationarity if a considerable amount of individual units of the panel are stationary series. If interest lies in distinguishing stationary from unit root series within the panel, a completely different approach should be developed within the Bayesian framework.
This data has been obtained from the Penn World table of Alan Heston, Robert Summers and Bettina Aten, Center for International Comparisons of Production, Income and Prices at the University of Pennsylvania.
The interest rates data were obtained from the archive of J. Huston McCulloch of Ohio State University of USA (http://www.econ.ohio-state.edu/jhm/ts/ts.html).
References
Arellano, M.: Panel Data Econometrics. Oxford University Press, Oxford (2003)
Baltagi, B.H., Kao, C.: Nonstationary panels, cointegration in panels and dynamic panels, a survey. In: Baltagi, B. (ed.) Nonstationary Panels, Panel Cointegration and Dynamic Panels. Advances in Econometrics, vol. 14, pp. 7–52 (2000)
Baltagi, B.H., Pirotte, A.: Panel data inference under spatial dependence. CPR Working Paper, 123 (2010)
Bernardo, J.M., Smith, A.F.M.: Bayesian Theory. Wiley Series in Applied Probability and Statistics (1994)
Box, J.E.P., Jenkins, G.M.: Time Series Analysis, Forecasting and Control. Holden Day, San Francisco (1976)
Breitung, J.: Nonparametric tests for unit roots and cointegration. J. Econom. 108, 343–363 (2002)
Breitung, J., Meyer, W.: Testing for unit roots using panel data: are wages on different bargaining levels cointegrated? Appl. Econ. 26, 353–361 (1994)
Breitung, J., Pesaran, M.H.: Unit roots and cointegration in panels. In: Matyas, Sevestre (eds.) The Econometrics of Panel Data, 3rd edn., pp. 279–322 (2008)
Campbell, J.Y., Lo, A.W., Mackinlay, A.C.: The Econometrics of Financial Markets. Princeton University Press, Princeton (1997)
Carriero, A., Favero, C., Kaminska, I.: Financial factors, macroeconomic information and the expectations theory of the term structure. J. Econom. 131, 339–359 (2006)
Chang, Y.: Bootstrap unit root tests in panels with cross-sectional dependence. J. Econom. 120, 263–293 (2004)
Dai, Q., Singleton, K.J.: Specification analysis of affine term structure models. J. Finance 55, 1943–1978 (2000)
DeJong, D.N., Whiteman, C.H.: Reconsidering ‘Trends and random walks in macroeconomic time series’. J. Monet. Econ. 28, 221–254 (1991)
DeJong, D.N., Whiteman, C.H.: Bayesian inference in dynamic equilibrium models: an application to the rational expectations model of the term structure. In: Berry, D.A., Chaloner, K.M., Geweke, J.K. (eds.) Bayesian Analysis in Statistics and Econometrics: Essays in Honor of Arnold Zellner, pp. 141–152. Wiley, New York (1996)
Elliot, G., Muller, U.: Tests for unit roots and the initial condition. Econometrica 71, 1269–1286 (2003)
Geweke, J.: Bayesian inference in econometric models using Monte Carlo integration. Econometrica 57, 1317–1339 (1989)
Hadri, K.: Testing for stationarity in heterogenous panel data. Econom. J. 3, 148–161 (2000)
Hamilton, J.D.: Time Series Analysis. Princeton University Press, Princeton (1994)
Harris, R.D.F., Tzavalis, E.: Inference for unit roots in dynamic panels where the time dimension is fixed. J. Econom. 91, 201–226 (1999)
Harris, R.D.F., Tzavalis, E.: Testing for unit roots in dynamic panels in the presence of a deterministic trend: re-examining the unit root hypothesis for real stock prices and dividends. Econom. Rev. 23, 149–166 (2004)
Hlouskova, J., Wagner, M.: The performance of panel unit root and stationary tests: results from a large scale simulation study. Econom. Rev. 25, 85–117 (2006)
Im, K.S., Pesaran, M.H., Shin, Y.: Testing for unit roots in heterogenous panels. J. Econom. 115, 53–74 (2003)
Jarocinski, M., Marcet, A.: Autoregressions and priors about the initial growth rates. Working paper, European Central Bank (2008)
Kang, W., Ferreira, M.: A Bayesian analysis of common trends and cycles on nonstationary time series panel data. Working Paper, Kent State University, New Philadelphia (2010)
Kapetanios, G.: Dynamic factor extraction of cross-sectional dependence in panel data unit root tests. J. Appl. Econom. 22, 313–338 (2007)
Kass, R.E., Raftery, A.E.: Bayes factors. J. Am. Stat. Assoc. 90, 773–795 (1995)
Levin, A., Lin, C., Chu, C.J.: Unit root tests in panel data: asymptotic and finite-sample properties. J. Econom. 108, 1–24 (2002)
Liu, J.S.: Monte Carlo Strategies in Scientific Computing. Springer, Berlin (2001)
Lubrano, M.: Testing for unit roots in a Bayesian framework. J. Econom. 69, 81–109 (1995)
Maddala, G.S., Kim, I.-M.: Unit Roots, Cointegration and Structural Change. Cambridge University Press, Cambridge (1999)
Maddala, G.S., Wu, S.: A comparative study of unit root tests with panel data and a new simple test. Oxf. Bull. Econ. Stat. 61, 631–652 (1999)
Marriott, J., Newbold, P.: The strength of evidence for unit autoregressive roots and structural breaks: A Bayesian perspective. J. Econom. 98, 1–25 (2000)
Meligkotsidou, L., Tzavalis, E., Vrontos, I.D.: A Bayesian analysis of unit roots and structural breaks on the level the trend and the variance of the autoregressive model error term. Econom. Rev. 30, 208–249 (2011)
Murray, C., Papell, D.H.: The purchasing power parity is worse than you think. Empir. Econ. 30, 783–790 (2005)
Moon, H.R., Perron, B.: Testing for a unit root in panels with dynamic factors. J. Econom. 122, 81–126 (2004)
Moon, H.R., Perron, B.: Testing for unit roots in panels with dynamic factors. J. Econom. 122, 81–126 (2005)
O’Connell, P.: The overvaluation of purchasing power parity. J. Int. Econ. 44, 1–19 (1998)
Papell, D.: The panel purchasing power parity puzzle. J. Money Credit Bank. 38, 447–467 (2006)
Pesaran, H.M.: A simple panel unit root test in the presence of cross-section dependence. J. Appl. Econom. 22, 265–312 (2007)
Phillips, P.C.B.: To criticize the critics: an objective Bayesian analysis of stochastic trends. J. Appl. Econom. 6, 333–364 (1991)
Phillips, P.C.B., Moon, H.R.: Linear regression limit theory for nonstationary panel data. Econometrica 67, 1057–1111 (1999)
Phillips, P.C.B., Sul, D.: Dynamic panel estimation and homogeneity testing under cross section dependence. Econom. J. 6, 217–259 (2003)
Rose, A.K.: Is the real interest rate stable? J. Finance 43, 1095–1112 (1988)
Sarno, L., Taylor, M.P.: Purchasing power parity and the real exchange rate. IMF Staff Pap. 49, 65–105 (2002)
Schotman, P.C., van Dijk, H.: A Bayesian analysis of the unit root in real exchange rates. J. Econom. 49, 195–238 (1991a)
Schotman, P.C., van Dijk, H.: On Bayesian routes to unit roots. J. Appl. Econom. 6, 387–401 (1991b)
Sims, C.A.: Bayesian scepticism on unit root econometrics. J. Econ. Dyn. Control 12, 463–474 (1988)
Taylor, A.M.: A century of purchasing power parity. Rev. Econ. Stat. 84, 139–150 (2002)
Tzavalis, E., Wickens, M.R.: Explaining the failures of the term spread models of the rational expectations hypothesis of the term structure. J. Money Credit Bank. 29, 364–380 (1997)
Uhlig, H.: On Jeffreys prior when using the exact likelihood function. Econom. Theory 10, 633–644 (1994)
Zivot, E.: A Bayesian analysis of the unit root hypothesis within an unobserved components model. Econom. Theory 10, 552–578 (1994)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Meligkotsidou, L., Tzavalis, E. & Vrontos, I.D. A Bayesian method of distinguishing unit root from stationary processes based on panel data models with cross-sectional dependence. Stat Comput 24, 297–315 (2014). https://doi.org/10.1007/s11222-012-9371-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-012-9371-3