Journal of Statistical Theory and Practice

, Volume 7, Issue 3, pp 515–536 | Cite as

A New Procedure for the Behrens-Fisher Problem that Guarantees Confidence Levels

  • A. J. HayterEmail author


This article considers the Behrens-Fisher problem of comparing two independent normal means without making any assumptions about the two unknown population variances. It is known that the natural statistic for this problem has a distribution that depends upon the ratio of the two variances, and that inferences can be made by using a t-distribution with degrees of freedom equal to the minimum of the degrees of freedom of the two variance estimates. This is the limiting distribution of the statistic as the variance ratio tends to zero or one. In this article a new procedure is developed that simultaneously provides inferences on both the means and the variance ratio. This procedure bounds the variance ratio away from zero and infinity, so it can allow inferences on the means that are sharper than those provided by the limiting distribution. Some numerical illustrations of the advantage provided by this new procedure are presented.


Normal distribution Pairwise comparison Unequal variances Confidence interval Confidence set Test procedure p-Value Acceptance set Beta distribution Behrens-Fisher 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Behrens, W.V. 1929. Ein beitrag zur Fehlerberechnung bei wenigen Beobachtungen (A contribution to error estimation with few observations). Landwirts. Jahrbcher, 68, 80737.Google Scholar
  2. Chang, C. H., and N. Pal. 2008. A revisit to the Behrens-Fisher problem: Comparison of five test methods. Commun. Stat. Simulation Comput., 37(6), 1064–1085.MathSciNetCrossRefGoogle Scholar
  3. Fisher, R. A. 1935. The fiducial argument in statistical inference. Ann. Eugenics, 8, 391–398.CrossRefGoogle Scholar
  4. Huang, P., B. C. Tilley, R. F. Woolson, and S. Lipsitz. 2005. Adjusting O’Brien’s test to control type I error for the generalized nonparametric Behrens-Fisher problem. Biometrics, 61(2), 532–539.MathSciNetCrossRefGoogle Scholar
  5. Hsu, P. L. 1938. Contributions to the theory of “Student’s” t-test as applied to the problem of two samples. In Statistical research memoris, Vol. II, 1–24. London, UK, Department of Statistics, University College.Google Scholar
  6. Kabaila, P., and J. Tuck. 2008. Confidence intervals utilizing prior information in the Behrens-Fisher problem. Austr. N. Z. J. Stat., 50(4), 309–328.MathSciNetCrossRefGoogle Scholar
  7. Mickey, M. R., and M. B. Brown. 1966. Bounds on the distribution functions of the Behrens-Fisher statistic. Ann. Math. Stat., 37, 639–642.MathSciNetCrossRefGoogle Scholar
  8. Neubert, K., and E. Brunner. 2007. A studentized permutation test for the non-parametric Behrens-Fisher problem. Comput. Stat. Data Anal., 51(10), 5192–5204.MathSciNetCrossRefGoogle Scholar
  9. Ramos, P. D., and D. F. Ferreira. 2010. A Bayesian solution to the multivariate Behrens-Fisher problem. Comput. Stat. Data Anal., 54(6), 1622–1633.MathSciNetCrossRefGoogle Scholar
  10. Satterthwaite, F. E. 1946. An approximate distribution of estimates of variance components. Biometrics Bull., 2, 110–114.CrossRefGoogle Scholar
  11. Scheffe, H. 1970. Practical solutions of the Behrens-Fisher problem. J. Am. Stat. Assoc., 65, 1501–1508.MathSciNetzbMATHGoogle Scholar
  12. Welch, B. L. 1947. The generalization of “Student’s” problem when several different population variances are involved. Biometrika, 34, 28–35.MathSciNetzbMATHGoogle Scholar
  13. Welch, B. L. 1949. Further note on Mrs. Aspin’s tables and on certain approximations to the tabled function. Biometrika, 36, 293–296.MathSciNetzbMATHGoogle Scholar

Copyright information

© Grace Scientific Publishing 2013

Authors and Affiliations

  1. 1.Department of Business Information and AnalyticsUniversity of DenverDenverUSA

Personalised recommendations