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

Hierarchically well formulated model...

Cristol, Stephen Michael.

Linked to FindBook

Google Book

Amazon

博客來

Hierarchically well formulated models in epidemiology.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Hierarchically well formulated models in epidemiology./
Author:	Cristol, Stephen Michael.
Description:	87 p.
Notes:	Source: Dissertation Abstracts International, Volume: 65-08, Section: B, page: 3971.
Contained By:	Dissertation Abstracts International65-08B.
Subject:	Health Sciences, Public Health. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3142141
ISBN:	0496891365

Hierarchically well formulated models in epidemiology.
Cristol, Stephen Michael.

Hierarchically well formulated models in epidemiology. - 87 p.

Source: Dissertation Abstracts International, Volume: 65-08, Section: B, page: 3971.

Thesis (Ph.D.)--Emory University, 2004.

The dissertation begins with a series of definitions and examples to build a foundation for a formal examination of hierarchically well formulated models. Generalized and practical effects are defined progressively for didactic purposes. Generalized effects are helpful in studying modeling because their computation is independent of the type of model being used. The practical effect is related to traditional effect measures. A coding dependence in non-hierarchically well formulated models is illustrated through the analysis of a data set using both an hierarchically and a non-hierarchically well formulated model and three different coding schemes. The demonstration serves as a proof by counter example that the generalized effects of non-hierarchically well formulated models are not invariant to coding. The demonstration is repeated with a different data set to further illustrate that there is no coding scheme for non-hierarchically well formulated models that will always produce the generalized effects determined using a hierarchically well formulated model. It is then proven that the generalized effects and practical effect for hierarchically well formulated models are invariant to the choice of coding for the data. A geometric interpretation of the findings is offered as a tangible explanation of why hierarchically well formulated models are important. Suggestions are given on how to modify modeling strategy to incorporate these results.

ISBN: 0496891365Subjects--Topical Terms:

1017659
Health Sciences, Public Health.

Hierarchically well formulated models in epidemiology.
LDR:02326nmm 2200277 4500 001 1839592
005 20050624104420.5
008 130614s2004 eng d
020 $a 0496891365
035 $a (UnM)AAI3142141
035 $a AAI3142141
040 $a UnM $c UnM
100 1 $a Cristol, Stephen Michael. $3 1927972
245 1 0 $a Hierarchically well formulated models in epidemiology.
300 $a 87 p.
500 $a Source: Dissertation Abstracts International, Volume: 65-08, Section: B, page: 3971.
500 $a Adviser: David G. Kleinbaum.
502 $a Thesis (Ph.D.)--Emory University, 2004.
520 $a The dissertation begins with a series of definitions and examples to build a foundation for a formal examination of hierarchically well formulated models. Generalized and practical effects are defined progressively for didactic purposes. Generalized effects are helpful in studying modeling because their computation is independent of the type of model being used. The practical effect is related to traditional effect measures. A coding dependence in non-hierarchically well formulated models is illustrated through the analysis of a data set using both an hierarchically and a non-hierarchically well formulated model and three different coding schemes. The demonstration serves as a proof by counter example that the generalized effects of non-hierarchically well formulated models are not invariant to coding. The demonstration is repeated with a different data set to further illustrate that there is no coding scheme for non-hierarchically well formulated models that will always produce the generalized effects determined using a hierarchically well formulated model. It is then proven that the generalized effects and practical effect for hierarchically well formulated models are invariant to the choice of coding for the data. A geometric interpretation of the findings is offered as a tangible explanation of why hierarchically well formulated models are important. Suggestions are given on how to modify modeling strategy to incorporate these results.
590 $a School code: 0665.
650 4 $a Health Sciences, Public Health. $3 1017659
650 4 $a Biology, Biostatistics. $3 1018416
690 $a 0573
690 $a 0308
710 2 0 $a Emory University. $3 1017429
773 0 $t Dissertation Abstracts International $g 65-08B.
790 1 0 $a Kleinbaum, David G., $e advisor
790 $a 0665
791 $a Ph.D.
792 $a 2004
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3142141