BAYESIAN ESTIMATES OF DISEASE MAPS - HOW IMPORTANT ARE PRIORS

被引：176

作者：

BERNARDINELLI, L ^{[1
]}

CLAYTON, D ^{[1
]}

MONTOMOLI, C ^{[1
]}

机构：

[1] INST PUBL HLTH,MRC,BIOSTAT UNIT,CAMBRIDGE CB2 2SR,ENGLAND

来源：

STATISTICS IN MEDICINE | 1995年 / 14卷 / 21-22期

关键词：

D O I：

10.1002/sim.4780142111

中图分类号：

Q [生物科学];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

In the fully Bayesian (FB) approach to disease mapping the choice of the hyperprior distribution of the dispersion parameter is a key issue. In this context we investigated the sensitivity of the rate ratio estimates to the choice of the hyperprior via a simulation study. We also compared the performance of the FB approach to mapping disease risk to the conventional approach of mapping maximum likelihood (ML) estimates and p-values. The study was modelled on the incidence data of insulin dependent diabetes mellitus (IDDM) as observed in the communes of Sardinia.

引用

页码：2411 / 2431

页数：21

共 20 条

[1] EMPIRICAL BAYES VERSUS FULLY BAYESIAN-ANALYSIS OF GEOGRAPHICAL VARIATION IN DISEASE RISK [J].

BERNARDINELLI, L ;

MONTOMOLI, C .

STATISTICS IN MEDICINE, 1992, 11 (08) :983-1007

[2]

BERNARDINELLI L, 1994, ATLAS CANCER MORTALI

[3] BAYESIAN IMAGE-RESTORATION, WITH 2 APPLICATIONS IN SPATIAL STATISTICS [J].

BESAG, J ;

YORK, J ;

MOLLIE, A .

ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS, 1991, 43 (01) :1-20

[4]

BIGGERI A, 1995, J ITALIAN STATISTICA, V2, P251

[5]

BRESLOW NE, 1993, J AM STAT ASSOC, V88, P421

[6]

BRESLOW NE, 1984, APPL STAT-J ROY ST C, V33, P38

[7] EMPIRICAL BAYES ESTIMATES OF AGE-STANDARDIZED RELATIVE RISKS FOR USE IN DISEASE MAPPING [J].

CLAYTON, D ;

KALDOR, J .

BIOMETRICS, 1987, 43 (03) :671-681

[8]

CLAYTON DG, 1993, SMALL AREA STUDIES G

[9]

CLAYTON DG, 1994, INT J EPIDEMIOL, V22, P1193

[10]

CLAYTON DG, 1989, 14TH P INT BIOM C, P201

← 1 2 →