Abstract
The goal of this paper is to describe the quality of different two-sample bootstrap and permutation tests for comparing variances. It is thereby inter alia shown that studentized resampling versions of the classical F-ratio test are asymptotically effective in a general nonparametric setting. This means that there is asymptotic no loss of power under contiguous alternatives. Moreover it is indicated that these tests are asymptotically consistent for fixed alternatives.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Billingsley P (1968) Convergence of probability measures. Wiley, New York
Boos DD, Brownie C (1989) Bootstrap methods for testing homogeneity of variances. Technometrics 31(1):69–82
Boos DD, Brownie C (2004) Comparing variances and other measures of dispersion. Stat Sci 19(4):571–578
Boos DD, Janssen P, Veraverbeke N (1989) Resampling from centered data in the two-sample problem. J Stat Plan Inference 21(3):327–345
Del Barrio E, Janssen A, Matrán C (2009) Resampling schemes with low resampling intensity and their applications in testing hypotheses. J Stat Plan Inference 139:184–202
Edgington ES, Onghena P (2007) Randomization tests. Statistics: textbooks and monographs. Chapman & Hall, Boca Raton
Good P (2005) Permutation, parametric and bootstrap tests of hypotheses, 3rd edn. Springer series in statistics. Springer, New York
Hájek J, Šidak Z, Sen PK (1999) Theory of rank tests. Academic Press, San Diego
Hayes AF (1997) Cautions in testing variance equality with randomization tests. J Stat Comput Simul 59:25–31
Hayes AF (2000) Randomization tests and the homoscedasticity assumption when comparing group means. Anim Behav 59:653–656
Janssen A (1997) Studentized permutation tests for non-i.i.d. hypotheses and the generalized Behrens–Fisher problem. Stat Probab Lett 36:9–21
Janssen A (2005) Resampling Student’s t-type statistics. Ann Inst Stat Math 57:507–529
Janssen A, Pauls T (2003a) How do bootstrap and permutation tests work? Ann Stat 31:768–806
Janssen A, Pauls T (2003b) A Monte Carlo comparison of studentized bootstrap and permutation tests for heteroscedastic two-sample problems. Comput Stat 20(3):369–383
Janssen A, Pauly M (2009). Asymptotics and effectiveness of conditional tests with applications to randomization tests. Technical report, University of Duesseldorf
Johnson NL, Kotz S (1970) Distributions in statistics: continuous univariate distributions, vols 1, 2. Houghton Mifflin, New York
Lehmann EL, Romano JP (2005) Testing statistical hypotheses. Springer, New York
Manly BFJ (1997) Randomization, bootstrap and Monte Carlo methods in biology. Chapman & Hall, London
Neuhaus G (1993) Conditional rank tests for the two-sample problem under random censorship. Ann Stat 21:1760–1779
Romano JP (1990) On the behavior of randomization tests without a group invariance assumption. J Am Stat A 85:686–692
Sakaori F (2002) Permutation test for equality of correlation coefficients in two populations. Commun Stat, Simul Comput 31:641–651
Van der Vaart AW (1998) Asymptotic statistics. Cambridge University Press, Cambridge
Zhang J, Boos DD (1993) Testing hypotheses about covariance matrices using bootstrap methods. Commun Stat, Theory Methods 22(3):723–739
Zhu L-X, Ng KW, Jing P (2002) Resampling methods for homogeneity tests of covariance matrices. Stat Sin 12(3):769–783
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Domingo Morales.
This article has been developed while the author was a fellow of the ‘Gründerstiftung for promotion of research of young academics’ at the Heinrich-Heine Universität Düsseldorf.
Rights and permissions
About this article
Cite this article
Pauly, M. Discussion about the quality of F-ratio resampling tests for comparing variances. TEST 20, 163–179 (2011). https://doi.org/10.1007/s11749-010-0194-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11749-010-0194-2
Keywords
- Heterogeneous null distributions
- Permutation tests
- Bootstrap tests
- Power
- Studentized statistics
- Two-sample tests