Analysis of Dispersion Effects in Fractional Factorial Two-Level Designs

  • Steffen Uhlig
Conference paper


In factorial designs that are used in the development of new products or in other applications the response Y frequently has unequal variances depending on the factor levels. The response can be written as a function Y = f(x), where x = (t′, Z′), entails all factors influencing Y. t denotes the treatment, i. e. the levels of the controllable factors (design factors), while Z denotes the levels of the (non-observable) random factors. It is assumed that all components of Z are stochastically independent.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Box, G. E. P and Meyer, R. D. [1986], Dispersion effects from fractional designs, Technometrics 30, 1–17CrossRefGoogle Scholar
  2. Davidian, M. and Carroll, R. J. [1987], Variance function estimation. Journal of the American Statistical Association 82, 1079–1091CrossRefGoogle Scholar
  3. Ghosh, S. and Lagergren, E. S. [1990], Dispersion models and estimation of dispersion effects in replicated factorial experiments. Journal of statistical planning and inference 26, 253–262CrossRefGoogle Scholar
  4. Rao, C. R. [1979], MINQE theory and its relation to ML and MML estimation of variance components. Sankhya 41, 138–153Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  • Steffen Uhlig
    • 1
  1. 1.Institut für Statistik und ÖkonometrieFreie Universität BerlinBerlin 33Germany

Personalised recommendations