Skip to main content
placeholder image

Playing safe with misweighted means

Journal Article


Abstract


  • This article concludes that Student’s t statistic is a forgiving or safe statistic in the presence of unequal variances. More generally, for every weighted mean (Equation presented), there is a corresponding weighted sum of deviations squared Sw, 02, which together with the mean defines a safe Student-like test statistic. These results imply that the choice of weights in (Equation presented) need not be all that precise, provided a compensation is made via Sw, 0. Such is the case in the example considered. © 1982 Taylor & Francis Group, LLC.

Publication Date


  • 1982

Citation


  • Cressie, N. (1982). Playing safe with misweighted means. Journal of the American Statistical Association, 77(380), 754-759. doi:10.1080/01621459.1982.10477882

Scopus Eid


  • 2-s2.0-3042603168

Web Of Science Accession Number


Start Page


  • 754

End Page


  • 759

Volume


  • 77

Issue


  • 380

Abstract


  • This article concludes that Student’s t statistic is a forgiving or safe statistic in the presence of unequal variances. More generally, for every weighted mean (Equation presented), there is a corresponding weighted sum of deviations squared Sw, 02, which together with the mean defines a safe Student-like test statistic. These results imply that the choice of weights in (Equation presented) need not be all that precise, provided a compensation is made via Sw, 0. Such is the case in the example considered. © 1982 Taylor & Francis Group, LLC.

Publication Date


  • 1982

Citation


  • Cressie, N. (1982). Playing safe with misweighted means. Journal of the American Statistical Association, 77(380), 754-759. doi:10.1080/01621459.1982.10477882

Scopus Eid


  • 2-s2.0-3042603168

Web Of Science Accession Number


Start Page


  • 754

End Page


  • 759

Volume


  • 77

Issue


  • 380