Skip to main content

A Levene-type test of homogeneity of variances against ordered alternatives

Abstract

Investigations focusing on differences in scale parameters across multiple samples appear in various scientific fields, e.g., when assessing measurement precision. In such cases a natural order of groups often suggests itself, which can be exploited by tests involving ordered alternatives. We propose multiple contrast tests with order-restriction based on robust Levene residuals as a powerful approach to detect scale differences between several groups. The proposed method is applied to a real data set and also extensive simulation is conducted to assess power and robustness against non-normality.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

References

  • Agresti A, Coull BA (1998) Order-restricted inference for monotone trend alternatives in contingency tables. Comput Stat Data Anal 28:139–155

    Article  MATH  Google Scholar 

  • Blackman MJ, Stein GJ, Vandiver PB (1993) The standardization hypothesis and ceramic mass production: technological, compositional and metric indexes of craft specialization at Tell Leilan, Syria. Am Antiq 58:60–80

    Article  Google Scholar 

  • Bretz F (2006) An extension of the Williams trend test to general unbalanced linear models. Comp Stat Data Anal 50:1735–1748

    Article  MATH  MathSciNet  Google Scholar 

  • Bretz F, Hothorn LA (2003) Statistical analysis of monotone or non-monotone dose-response data from in vitro toxicological assays. Altern Lab Anim: ATLA 31:81–96

    Google Scholar 

  • Brown MB, Forsythe AB (1974) Robust tests for equality of variances. J Am Stat Assoc 69:364–367

    Article  MATH  Google Scholar 

  • Cahoy DO (2010) A bootstrap test for equality of variances. Comput Stat Data Anal 54:2306–2316

    Article  MATH  MathSciNet  Google Scholar 

  • Dilba G, Schaarschmidt F, Hothorn LA (2007) Inferences for ratios of normal means. R News 7:20–23

    Google Scholar 

  • Djira GD, Hothorn LA (2009) Detecting relative changes in multiple comparisons with an overall mean. J Qual Technol 41:60–65

    Google Scholar 

  • Djira GD, Hasler M, Gerhard D, Schaarschmidt F (2012) mratios: Inferences for ratios of coefficients in the general linear model. R package version 1.3.17, available at http://cran.r-project.org/package=mratios

  • Dunnett C (1955) A multiple comparison procedure for comparing several treatments with a control. J Am Stat Assoc 50:1096–1121

    Article  MATH  Google Scholar 

  • Fujino Y (1979) Tests for the homogeneity of a set of variances against ordered alternatives. Biometrika 66:133–139

    Article  MATH  MathSciNet  Google Scholar 

  • Gastwirth JL, Gel YR, Miao W (2009) The impact of Levene’s test of equality of variances on statistical theory and practice. Stat Sci 24:343–360

    Article  MathSciNet  Google Scholar 

  • Gastwirth JL, Gel YR, Hui WLW, Lyubchich V, Miao W, Noguchi K (2013) lawstat: An R package for biostatistics, public policy, and law. R package version 2.4.1, available at http://cran.r-project.org/package=lawstat

  • Genz A, Bretz F (2009) Computation of multivariate normal and \(t\) probabilities. Springer, Heidelberg

    Book  MATH  Google Scholar 

  • Genz A, Bretz F, Miwa T, Mi X, Leisch F, Scheipl F, Hothorn T (2012) mvtnorm: Multivariate normal and \(t\) distributions. R package version 0.9-9992, available at http://cran.r-project.org/package=mvtnorm

  • Gould SJ (1988) Trends as changes in variance: a new slant on progress and directionality in evolution. J Paleontol 62:319–329

    Google Scholar 

  • Hayter AJ (1990) A one-sided studentized range test for testing against a simple ordered alternative. J Am Stat Assoc 85:778–785

    MATH  MathSciNet  Google Scholar 

  • Hirotsu C, Yamamoto S, Hothorn LA (2011) Estimating the dose-response pattern by the maximal contrast type test approach. Stat Biopharm Res 3:40–53

    Article  Google Scholar 

  • Hothorn LA, Libiger O, Gerhard D (2012) Model-specific tests on variance heterogeneity for detection of potentially interacting genetic loci. BMC Genet 13:article 59

  • Hothorn T, Bretz F, Westfall P (2008) Simultaneous inference in general parametric models. Biom J 50:346–363

    Article  MathSciNet  Google Scholar 

  • Keyes TK, Levy MS (1997) Analysis of Levene’s test under design imbalance. J Educ Behav Stat 22:227–236

    Article  Google Scholar 

  • Konietschke F, Hothorn LA (2012) Evaluation of toxicological studies using a nonparametric Shirley-type trend test for comparing several dose levels with a control group. Stat Biopharm Res 4:14–27

    Article  Google Scholar 

  • Konietschke F, Hothorn LA, Brunner E (2012) Rank-based multiple test procedures and simultaneous confidence intervals. Electron J Stat 6:738–759

    Article  MATH  MathSciNet  Google Scholar 

  • Konietschke F, Bösiger S, Brunner E, Hothorn LA (2013) Are multiple contrast tests superior to the ANOVA? Int J Biostat 9:1–11

    Article  Google Scholar 

  • Kvamme KL, Stark MT, Longacre WA (1996) Alternative procedures for assessing standardization in ceramic assemblages. Am Antiq 61:116–126

    Article  Google Scholar 

  • Levene H (1960) Robust tests for equality of variances. In: Olkin I (ed) Contributions to probability and statistics. Stanford University Press, Palo Alto, pp 278–292

    Google Scholar 

  • Loh WY (1987) Some modifications of Levene’s test of variance homogeneity. J Stat Comput Simul 28:213–226

    Article  Google Scholar 

  • Marcus R (1982) Some results on simultaneous confidence-intervals for monotone contrasts in one-way ANOVA model. Commun Stat A-Theory Methods 11:615–622

    Article  Google Scholar 

  • Marozzi M (2011) Levene type tests for the ratio of two scales. J Stat Comput Simul 81:815–826

    Article  MATH  MathSciNet  Google Scholar 

  • Marozzi M (2012) A distribution free test for the equality of scales. Commun Stat-Simul Comput 41:878–889

    Article  MATH  MathSciNet  Google Scholar 

  • Marozzi M (2013) Nonparametric simultaneous tests for location and scale testing: a comparison of several methods. Commun Stat-Simul Comput 42:1298–1317

    Article  MATH  MathSciNet  Google Scholar 

  • McDermott MP (1999) Generalized orthogonal contrast tests for homogeneity of ordered means. Can J Stat-Rev Can Stat 27:457–470

    Article  MATH  Google Scholar 

  • Neuhäuser M, Hothorn LA (2000) Parametric location-scale and scale trend tests based on Levene’s transformation. Comput Stat Data Anal 33:189–200

    Article  MATH  Google Scholar 

  • Neuhäuser M, Leuchs AK, Ball D (2011) A new location-scale test based on a combination of the ideas of Levene and Lepage. Biom J 53:525–534

    Article  MATH  MathSciNet  Google Scholar 

  • Noguchi K, Gel YR (2010) Combination of Levene-type test and a finite-intersection method for testing equality of variances against ordered alternatives. J Nonparametr Stat 22:897–913

    Article  MATH  MathSciNet  Google Scholar 

  • O’Neill ME, Mathews K (2000) A weighted least squares approach to Levene’s test of homogeneity of variance. Aust N Z J Stat 42:81–100

    Article  MATH  MathSciNet  Google Scholar 

  • Parra-Frutos I (2009) The behaviour of the modified Levene’s test when data are not normally distributed. Comput Stat 24:671–693

    Article  MATH  MathSciNet  Google Scholar 

  • Parra-Frutos I (2013) Testing homogeneity of variances with unequal samples sizes. Comput Stat 28:1269–1297

    Article  MathSciNet  Google Scholar 

  • Peng J, Lee CIC, Liu L, Jiang J (2006) Orthogonal multiple contrast test for testing monotone on the average. Stat Probab Lett 76:161–168

    Article  MATH  MathSciNet  Google Scholar 

  • Development Core Team R (2012) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna

  • Rublík F (2010) A note on simultaneous confidence intervals for ratio of variances. Commun Stat-Theory Methods 39:1038–1045

    Article  MATH  Google Scholar 

  • Singh P (2006) Simultaneous confidence intervals for the successive ratios of scale parameters. J Stat Plan Inf 136:1007–1019

    Article  MATH  Google Scholar 

  • Singh P, Gill AN, Kumar N (2009) A test of homogeneity of several variances against tree ordered alternative. Stat Probab Lett 79:2315–2320

    Article  MATH  MathSciNet  Google Scholar 

  • Singh P, Goyal A, Gill AN (2010) A note on comparing several variances with a control variance. Stat Probab Lett 80:1995–2002

    Article  MATH  MathSciNet  Google Scholar 

  • Williams DA (1971) A test for differences between treatment means when several dose levels are compared with a zero dose control. Biometrics 27:103–117

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Philip Pallmann.

Additional information

The work of the second author (LAH) was partly supported by the German Science Foundation grant DFG-HO1687. The authors wish to thank the editor and two anonymous referees for hints and suggestions that helped improve the manuscript.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (R 2 KB)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Pallmann, P., Hothorn, L.A. & Djira, G.D. A Levene-type test of homogeneity of variances against ordered alternatives. Comput Stat 29, 1593–1608 (2014). https://doi.org/10.1007/s00180-014-0508-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00180-014-0508-z

Keywords

  • Levene residuals
  • Heterogeneous variances
  • Trend test
  • Multiple contrasts
  • Order-restricted inference