東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Locally D-optimal designs for nonlin...

Li, Gang.

Linked to FindBook

Google Book

Amazon

博客來

Locally D-optimal designs for nonlinear models with minimal number of support points.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Locally D-optimal designs for nonlinear models with minimal number of support points./
Author:	Li, Gang.
Description:	117 p.
Notes:	Source: Dissertation Abstracts International, Volume: 66-07, Section: B, page: 3781.
Contained By:	Dissertation Abstracts International66-07B.
Subject:	Statistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3183685
ISBN:	0542250071

Locally D-optimal designs for nonlinear models with minimal number of support points.
Li, Gang.

Locally D-optimal designs for nonlinear models with minimal number of support points. - 117 p.

Source: Dissertation Abstracts International, Volume: 66-07, Section: B, page: 3781.

Thesis (Ph.D.)--University of Illinois at Chicago, 2005.

Nonlinear models arise naturally in many important applications, such as biological, pharmacological and biomedical statistics. The goal of this research is to identify optimal and highly efficient designs for such experiments. Although the optimal design theory for linear models has been developed and applied for many years, there are much less results for nonlinear models.

ISBN: 0542250071Subjects--Topical Terms:

517247
Statistics.

Locally D-optimal designs for nonlinear models with minimal number of support points.
LDR:02301nmm 2200301 4500 001 1814496
005 20060526065256.5
008 130610s2005 eng d
020 $a 0542250071
035 $a (UnM)AAI3183685
035 $a AAI3183685
040 $a UnM $c UnM
100 1 $a Li, Gang. $3 1278067
245 1 0 $a Locally D-optimal designs for nonlinear models with minimal number of support points.
300 $a 117 p.
500 $a Source: Dissertation Abstracts International, Volume: 66-07, Section: B, page: 3781.
500 $a Chairperson: Dibyen Majumdar.
502 $a Thesis (Ph.D.)--University of Illinois at Chicago, 2005.
520 $a Nonlinear models arise naturally in many important applications, such as biological, pharmacological and biomedical statistics. The goal of this research is to identify optimal and highly efficient designs for such experiments. Although the optimal design theory for linear models has been developed and applied for many years, there are much less results for nonlinear models.
520 $a It follows from Caratheodory's theorem that any D-optimal design for a k-parameter model is supported on at least k points. The characterization of minimally supported (saturated) designs has been a main focus of the optimal design literature.
520 $a In this dissertation, we first establish several new sufficient conditions to characterize the classes of models for which the (locally) D-optimal designs are minimally supported. We apply these conditions to provide alternative proofs for existing results as well as new results for a variety of models. These models include compartmental models, logistic models and Gompertz models.
520 $a For some models that we cannot verify these sufficient conditions, we explore methods to narrow the search for optimal designs. In addition, we investigate the efficiency of the equally spaced and uniform designs for some models. Both bounded and unbounded design spaces are considered. Throughout the dissertation we work in the "approximate theory" setup.
590 $a School code: 0799.
650 4 $a Statistics. $3 517247
690 $a 0463
710 2 0 $a University of Illinois at Chicago. $3 1020478
773 0 $t Dissertation Abstracts International $g 66-07B.
790 1 0 $a Majumdar, Dibyen, $e advisor
790 $a 0799
791 $a Ph.D.
792 $a 2005
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3183685