Abstract
The sup\(LM\) test for structural change is embedded into a permutation test framework for a simple location model. The resulting conditional permutation distribution is compared to the usual (unconditional) asymptotic distribution, showing that the power of the test can be clearly improved in small samples. Furthermore, the permutation test is embedded into a general framework that encompasses tools for binary and multivariate dependent variables as well as model-based permutation testing for structural change. It is also demonstrated that the methods can not only be employed for analyzing structural changes in time series data but also for recursive partitioning of cross-section data. The procedures suggested are illustrated using both artificial data and empirical applications (number of youth homicides, employment discrimination data, carbon flux in tropical forests, stock returns, and demand for economics journals).
Similar content being viewed by others
References
Andrews DWK (1991) Heteroskedasticity and autocorrelation consistent covariance matrix estimation. Econometrica 59:817–858
Andrews DWK (1993) Tests for parameter instability and structural change with unknown change point. Econometrica 61:821–856
Antoch J, Hušková M (2001) Permutation tests in change point analysis. Stat Probab Lett 53:37–46
Bergstrom TC (2001) Free labor for costly journals? J Econ Perspect 15:183–198
Bonal D, Bosc A, Ponton S, Goret JY, Burban B, Gross P, Bonnefond JM, Elbers J, Longdoz B, Epron D, Guehl JM, Granier A (2008) Impact of severe dry season on net ecosystem exchange in the neotropical rainforest of French Guiana. Glob Chang Biol 14(8):1917–1933
Boulesteix AL, Strobl C (2007) Maximally selected chi-squared statistics and non-monotonic associations: an exact approach based on two cutpoints. Comput Stat Data Anal 51(12):6295–6306
Brown RL, Durbin J, Evans JM (1975) Techniques for testing the constancy of regression relationships over time. J R Stat Soc B 37:149–163
Chambers JQ, Schimel JP, Nobre AD (2001) Respiration from coarse wood litter in central Amazon forests. Biogeochemistry 52(2):115–131
Chow GC (1960) Tests of equality between sets of coefficients in two linear regressions. Econometrica 28:591–605
Ernst MD (2004) Permutation methods: a basis for exact inference. Stat Sci 19(4):676–685
Fisher RA (1935) The design of experiments. Oliver and Boyd, Edinburgh
Freidlin B, Gastwirth JL (2000) Changepoint tests designed for the analysis of hiring data arising in employment discrimination cases. J Bus Econ Stat 18(3):315–322
Genz A (1992) Numerical computation of multivariate normal probabilities. J Comput Graph Stat 1:141–149
Hansen BE (1997) Approximate asymptotic \(p\) values for structural-change tests. J Bus Econ Stat 15:60–67
Hansen BE (2000) Testing for structural change in conditional models. J Econ 97:93–115
Hastie T, Tibshirani R, Friedman J (2009) The elements of statistical learning, 2nd edn. Springer, New York
Hothorn T, Lausen B (2003) On the exact distribution of maximally selected rank statistics. Comput Stat Data Anal 43(2):121–137
Hothorn T, Zeileis A (2008) Generalized maximally selected statistics. Biometrics 64:1263–1269
Hothorn T, Hornik K, van de Wiel MA, Zeileis A (2006a) A Lego system for conditional inference. Am Stat 60(3):257–263
Hothorn T, Hornik K, Zeileis A (2006b) Unbiased recursive partitioning: a conditional inference framework. J Comput Graph Stat 15(3):651–674
Hothorn T, Hornik K, van de Wiel MA, Zeileis A (2008) Implementing a class of permutation tests: the coin package. J Stat Softw 28(8):1–23. http://www.jstatsoft.org/v28/i08/
Hsu DA (1979) Detecting shifts of parameter in gamma sequences with applications to stock price and air traffic flow analysis. J Am Stat Assoc 74:31–40
Kennedy PE (1995) Randomization tests in econometrics. J Bus Econ Stat 13(1):85–94
Kirch C (2007) Block permutation principles for the change analysis of dependent data. J Stat Plan Inference 137:2453–2474
Kirch C, Steinebach J (2006) Permutation principles for the change analysis of stochastic processes under strong invariance. J Comput Appl Math 186(1):64–88
Krämer W, Sonnberger H (1986) The linear regression model under test. Physica, Heidelberg
Krämer W, Ploberger W, Alt R (1988) Testing for structural change in dynamic models. Econometrica 56(6):1355–1369
Lausen B, Schumacher M (1992) Maximally selected rank statistics. Biometrics 48:73–85
Ludbrook J, Dudley H (1998) Why permutation tests are superior to \(t\) and \(F\) tests in biomedical research. Am Stat 52(2):127–132
Luger R (2006) Exact permutation tests for non-nested non-linear regression models. J Econ 133:513–529
McCullagh P (2005) Exchangeability and regression models. In: Davison AC, Dodge Y, Wermuth N (eds) Celebrating statistics—papers in honour of Sir David Cox on his 80th birthday. Oxford University Press, Oxford, pp 89–115
Pesarin F (2001) Multivariate permutation tests: with applications to biostatistics. Wiley, Chichester
Piehl AM, Cooper SJ, Braga AA, Kennedy DM (2003) Testing for structural breaks in the evaluation of programs. Rev Econ Stat 85(3):550–558
Pitman EJG (1938) Significance tests which may be applied to samples from any populations: III. The analysis of variance test. Biometrika 29:322–335
Quandt RE (1960) Tests of the hypothesis that a linear regression obeys two separate regimes. J Am Stat Assoc 55:324–330
R Development Core Team (2012) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org/, ISBN 3-900051-07-0
Stahl C, Burban B, Goret JY, Bonal D (2011) Seasonal variations in stem CO\(_2\) efflux in the neotropical rainforest of French Guiana. Ann For Sci 68(4):771–782
Stock JH, Watson MW (1996) Evidence on structural instability in macroeconomic time series relations. J Bus Econ Stat 14:11–30
Stock JH, Watson MW (2007) Introduction to econometrics, 2nd edn. Addison-Wesley, Reading
Strasser H, Weber C (1999) On the asymptotic theory of permutation statistics. Math Methods Stat 8:220–250. Preprint available from http://epub.wu.ac.at/102/
Strobl C, Boulesteix AL, Augustin T (2007) Unbiased split selection for classification trees based on the Gini index. Comput Stat Data Anal 52:483–501
Zeileis A (2005) A unified approach to structural change tests based on ML scores, \(F\) statistics, and OLS residuals. Econ Rev 24(4):445–466
Zeileis A, Leisch F, Hornik K, Kleiber C (2002) strucchange: An R package for testing for structural change in linear regression models. J Stat Softw 7(2):1–38, http://www.jstatsoft.org/v07/i02/
Zeileis A, Hothorn T, Hornik K (2008) Model-based recursive partitioning. J Comput Graph Stat 17(2):492–514
Acknowledgments
The coarse woody debris respiration data were kindly provided by Lucy Rowland (School of GeoSciences, University of Edinburgh).
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
1.1 Proofs
The (asymptotic) distribution of the multivariate statistic \(Z = (Z_{\pi _1}, \dots , Z_{\pi _m})^\top \) is derived by embedding the statistic into the framework of Strasser and Weber (1999), as discussed in Hothorn et al. (2006a). More precisely, the test statistic \(\max _{\pi \in \Pi } Z_\pi \) considered above is the maximum-type statistic \(c_{\max }\) of Hothorn et al. (2006a) if the influence function \(h(Y) = Y\) is used for the observations \(Y_i\) and the transformation \(g(t) = (\mathbf{1 }_{[0, \pi _1]}(t), \dots , \mathbf{1 }_{[0, \pi _m]}(t))^\top \) is used for the associated timings \(t_i\). The transformation \(g\) used the indicator function \(\mathbf{1 }_I\) of the interval \(I\) and thus corresponds to a vector of indicators for the time up to the timings \(\pi _j\) (\(j = 1, \dots , k\)).
Using these transformations \(h(\cdot )\) and \(g(\cdot )\), the unstandardized test statistic \(T\) is in the notation of Hothorn et al. (2006a)
Under \(H_0\), given all permutations \(\sigma \in S\) of the observations \(Y_1, \dots , Y_n\), the unstandardized statistic has expectation
and each unstandardized statistic has variance
where the residual sum of squares is \(RSS_0 = \sum _{i = 1}^n (Y_i - \bar{Y})^2\). The equations directly follow from Eq. 7 in Strasser and Weber (1999).
Standardizing the vector of raw statistics \(T = (T_{\pi _1}, \dots , T_{\pi _m})^\top \) by their respective mean and standard deviation yields the vector of statistics \(Z = (Z_{\pi _1}, \dots , Z_{\pi _m})^\top \):
because of the following simple relationship between \(\bar{Y}_{1, \pi }\), \(\bar{Y}_{2, \pi }\) and \(\bar{Y}\):
Consequently, \(Z\) has zero mean and unit variance given all permutations \(\sigma \in S\). Similarly, the covariance between two elements of \(Z\), \(Z_\pi \) and \(Z_\tau \) say, is
Assuming that \(\pi < \tau \) and using the variance computed above, the correlation is thus
Given that we derived the first two moments of \(Z\) by embedding the statistic into the framework of Strasser and Weber (1999), the asymptotic normality of \(Z\) follows by application of their Theorem 2.3.
1.2 Power and size for autocorrelated series
To study the performance of the tests with dependent data, we use a simulation setup as in Sect. 2.3. The only difference is that the errors are now autocorrelated with \(\varrho = 0, 0.1, 0.2, 0.3, 0.5, 0.9\). The length of the time series considered is either very short (\(n = 10\)) or moderate (\(n = 50\)) and either there is no change (\(\delta = 0\)) or a large shift in the mean (\(\delta = 15\)). Five different versions of the tests are assessed: the unconditional and conditional test (\(\mathcal D _{\infty }\) and \(\mathcal D _{\sigma |Y}\), respectively) on the original data (as in Sect. 2.3), the unconditional test computed with a robust HAC covariance estimate and the unconditional and conditional test computed on the residuals of an AR(1) model (fitted by OLS).
Using the uncorrected tests on the original data (\(\mathcal D _{\infty }\) and \(\mathcal D _{\sigma |Y}\), respectively), it can be seen that size distortions occur for \(\varrho > 0\) even if there is no change (\(\delta = 0\)). For \(\varrho \) up to \(0.2\) these are still moderate but become very large afterwards and are even more pronounced for the conditional version of the test. However, in moderately large time series (\(n = 50\)), the problem can be remedied by either using a HAC correction or applying the tests to the AR(1) residuals. These three tests keep their size and have reasonable power with the conditional test having the highest power. However, none of the tests is able to distinguish between autocorrelation and a shift in the mean (with \(\delta = 15\)) if the time series is very short (\(n = 10\)) or if the autocorrelation is very high (\(\varrho \ge 0.7\)) and the length only moderate (\(n = 50\)).
Rights and permissions
About this article
Cite this article
Zeileis, A., Hothorn, T. A toolbox of permutation tests for structural change. Stat Papers 54, 931–954 (2013). https://doi.org/10.1007/s00362-013-0503-4
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-013-0503-4