Critical values for multiple structural change tests. (English) Zbl 1032.62064

J. Bai and P. Perron [Econometrica 66, 47-78 (1998; Zbl 1056.62523)] considered theoretical issues related to the limiting distribution of estimators and test statistics in the linear model with multiple structural changes. The asymptotic distributions of the tests depend on a trimming parameter \(\varepsilon\) and critical values were tabulated for \(\varepsilon=0.05\). As discussed by J. Bai and P. Perron [Multiple structural change models: a simulation analysis. Unpublished manuscript, Dpt. Economics, Boston Univ. (2000)], larger values of \(\varepsilon\) are needed to achieve tests with correct size in finite samples, when allowing for heterogeneity across segments or serial correlation in the errors. The aim of this paper is to supplement the set of critical values available with other values of \(\varepsilon\) to enable proper empirical applications. We provide response surface regressions valid for a wide range of parameters.


62J05 Linear regression; mixed models
62H12 Estimation in multivariate analysis
62H15 Hypothesis testing in multivariate analysis


Zbl 1056.62523
Full Text: DOI


[1] Andrews D. W. K., Econometrica 61 pp 821– (1993)
[2] DOI: 10.1016/0304-4076(94)01682-8 · Zbl 0834.62066 · doi:10.1016/0304-4076(94)01682-8
[3] Bai J., Econometric Theory 13 pp 315– (1997)
[4] Bai J., Econometrica 66 pp 47– (1998)
[5] 5. J. Bai, and P. Perron (2000 ); Multiple structural change models: a simulation analysis, unpublished manuscript, Department of Economics, Boston University.
[6] DOI: 10.1002/jae.659 · doi:10.1002/jae.659
[7] Garcia R., Review of Economics and Statistics 78 pp 111– (1996)
[8] Liu J., Statistica Sinica 7 pp 497– (1997)
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.