Abstract
This paper considers the problem of making inferences on a linear combination of means from independent normal distributions with unknown variances. While standard inference procedures require the assumption that the variances of the normal distributions are all equal or have known ratios, in this paper procedures are developed which allow the variances to take any unknown values. Bounds are developed for the confidence levels of confidence intervals, and for the sizes and p-values of hypothesis tests, which are valid for any values of the variances. These bounds are sharp in the sense that they are attained for certain limiting values of the variances. Applications to the one-way linear model are illustrated, and the Behrens-Fisher problem, which is a special case with just two means and for which the popular approximate Welch method can be employed, is also discussed. Some examples and simulations of the implementation of the procedures are provided, and comparisons are made with other procedures.
Similar content being viewed by others
References
Banerjee, S.K. (1960). Approximate confidence interval for linear functions of means of k-populations when the populations are not equal. Sankhya, 22, 357–258.
Behrens, W.V. (1929). Ein beitrag zur Fehlerberechnung bei wenigen Beobachtungen (translation: A contribution to error estimation with few observations). Landwirtschaftliche Jahrbcher, 68, 807–837.
Box, G.E.P., Hunter, W.G. and Hunter, J.S. (1978). Statistics for experimenters. Wiley.
Chang, C.H. & Pal, N. (2008). A revisit to the Behrens-Fisher problem: Comparison of five test methods. Comm. Statist. Simulation Comput., 37, 1064–1085.
Dalal, S.R. (1978). Simultaneous confidence procedures for univariate and multivariate Behrens-Fisher type problems. Biometrika, 65, 221–224.
Dobson, A.J. (1990). An introduction to generalized linear models. Chapman & Hall.
Ferguson, T.S. (1973). A Bayesian analysis of some nonparametric problems. Ann. Statist., 1, 209–230.
Fisher, R.A. (1935). The fiducial argument in statistical inference. Ann. Eugenics, 8, 391–398.
Hajek, J. (1962). Inequalities for the generalized Student’s distribution. Sel. Transl. Math. Statist. Prob., 2, 63–74.
Hsu, P.L. (1938). Contributions to the theory of “Student’s” t-test as applied to the problem of two samples. Statistical Research Memoirs, Vol. II. Department of Statistics, University College, London, pp. 1–24.
Huang, P., Tilley, B.C., Woolson, R.F. and Lipsitz, S. (2005). Adjusting O’Brien’s test to control type I error for the generalized nonparametric Behrens-Fisher problem. Biometrics, 61, 532–539.
Jensen, J.L.W.V. (1906). Sur les fonctions convexes et les inegalities entre les valeurs moyennes. Acta Math., 30, 175–193.
Jones, R.H. (1987). Serial correlation in unbalanced mixed models. Bull. Int. Stat. Inst., 52, 105–122.
Kabaila, P. and Tuck, J. (2008). Confidence intervals utilizing prior information in the Behrens-Fisher problem. Aust. N. Z. J. Stat., 50, 309–328.
Lawton, W.H. (1965). Some inequalities for central and non-central distributions. Ann. Math. Statist., 36, 1521–1524.
Lord, E. (1947). The use of range in place of the standard deviation in the t-test. Biometrika, 34, 41–67.
Lord, E. (1950). Power of the modified t-test (u-test) based on range. Biometrika, 37, 64–77.
Mickey, M.R. and Brown, M.B. (1966). Bounds on the distribution functions of the Behrens-Fisher statistic. Ann. Math. Stat., 37, 639–642.
Nel, D.G. and van der Merwe, C.A. (1986). A solution to the multivariate Behrens-Fisher problem. Comm. Statist. Theory Methods, 15, 3719–3735.
Neubert, K. and Brunner, E. (2007). A studentized permutation test for the non-parametric Behrens-Fisher problem. Comput. Statist. Data Anal., 51, 5192–5204.
Ramos, P.D. and Ferreir, D.F. (2010). A Bayesian solution to the multivariate Behrens-Fisher problem. Comput. Statist. Data Anal., 54, 1622–1633.
Ramsey, P.H. and Ramsey, P.P. (2008). Power of pairwise comparisons in the equal variance and unequal sample size case. Br. J. Math. Stat. Psychol., 61, 115–131.
Ramsey, P.H., Barrera, K., Hachimine-Semprebom, P. and Liu, C. (2011). Pairwise comparisons of means under realistic nonnormality, unequal variances, outliers and equal sample sizes. J. Stat. Comput. Simul., 81, 125–135.
Satterthwaite, F.E. (1946). An approximate distribution of estimates of variance components. Biometrics Bull., 2, 110–114.
Spjotvoll, E. (1972). Joint confidence intervals for all linear functions of means in ANOVA with unknown variances. Biometrika, 59, 684–685.
Tamhane, A.C. (1979). A comparison of procedures for multiple comparisons of means with unequal variances. J. Amer. Statist. Assoc., 74, 366, 471–480.
Welch, B.L. (1947). The generalization of “Student’s” problem when several different population variances are involved. Biometrika, 34, 28–35.
Welch, B.L. (1949). Further note on Mrs. Aspin’s tables and on certain approximations to the tabled function. Biometrika, 36, 293–296.
Wilks, S.S. (1962) Mathematical statistics. Wiley.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hayter, A.J. Inferences on Linear Combinations of Normal Means with Unknown and Unequal Variances. Sankhya A 76, 257–279 (2014). https://doi.org/10.1007/s13171-013-0041-0
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13171-013-0041-0
Keywords and phrases
- Normal distribution
- unequal variances
- confidence interval
- hypothesis test procedure
- p-value
- Dirichlet distribution
- Behrens-Fisher
- one-way layout
- Jensen’s inequality