C-Optimal Designs of Experiments for Estimation in Simplex Dispersion Model

• Activity Number: 432
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Mong-Na Lo Huang*+ and Hsiang-Ling Hsu
• Companies: National Sun Yat-Sen University and Academia Sinica
• Keywords: Elfing theorem ; proportional data ; prediction ; quantal estimation

Abstract

In this talk, the problem of determining an optimal experimental design for estimating linear combination of unknown parameters of a simplex dispersion model, based primary on Jørgensen (1997), is considered. The simplex dispersion model can be applied to proportional outcomes confined within the interval (0,1). C-optimal designs for accurate estimation of linear combination of parameters in the simplex dispersion model are characterized through the Elfving theorem and the corresponding equivalence theorem. Some numerical results are also presented. Examples about dose-response or psychometric studies are used to demonstrate that it would be more efficient to use c-optimal designs for estimation of certain quantities while conducting an experiment for proportional data.