An Optimum Design for First Order Fits to Correlated Responses

  • Bae, Wha-Soo (Department of Statistics Inje University, Kimhae, Kyongnam, 621-749, Korea.)
  • Published : 1996.12.01

Abstract

The aim of this paper is to find a suitable design which minimizes the expected discrepancy: in fitting a first order model fearing quadratic terms as bias where there are more than two correlated responses. Kim and Draper(1994) discussed about choosing a design for straight line fits to two correlated responses The general case with r responses is examined here and the result is applied to a specific case to help understandings.

Keywords

References

  1. Journal of American Statistical Associaion v.54 A basis for the selection of a response surface design Box, G. E. P.;Draper, N. R.
  2. Biometrika v.50 The choice of a second order rotatable designs Box, G. E. P.;Draper, N. R.
  3. Bayesian Inference in Statistical Analysis Box, G. E. P.;Tiao, G. C.
  4. Technical Report 302, Dept. of Stat. The analysis of multiresponse experiments: a review Khuri, A. I.
  5. Sttistica Sinica v.4 no.1 Choosing a design for straight line fits to two correlated responses Kim, W. B.;Draper, N. R.