IJPAM: Volume 38, No. 4 (2007)

GENERALIZED KIEFER AND WOLFOWITZ THEORY
IN D-OPTIMAL REGRESSION DESIGN

Qiang Guo$^1$, Jian-Guo Liu$^{2}$
$^1$School of Science
Dalian Nationalities University
Dalian, 116600, P.R. CHINA
$^2$ Institute of System Engineering
Dalian University of Technology
Dalian, 116024, P.R. CHINA
e-mail: liujg004@yahoo.com.cn


Abstract.In this paper, the D-optimal regression designs in heteroscedastic models are considered. The theory of Kiefer and Wolfowitz (KWT) is generated to the heteroscedastic models, called GKWT. GKWT can be used to vertify the optimality of the D-optimal regression designs in heteroscedastic models.

Received: June 20, 2007

AMS Subject Classification: 62K05

Key Words and Phrases: D-optimal design, approximate design, balanced incomplete block design

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2007
Volume: 38
Issue: 4