SINGULAR PRIOR DISTRIBUTIONS AND ILL-CONDITIONING IN BAYESIAN D-OPTIMAL DESIGN FOR SEVERAL NONLINEAR MODELS

SINGULAR PRIOR DISTRIBUTIONS AND ILL-CONDITIONING IN BAYESIAN D-OPTIMAL DESIGN FOR SEVERAL NONLINEAR MODELS

For Bayesian D-optimal design, we define a singular prior distribution for the model parameters as a prior distribution such that the determinant of the Fisher information matrix has a prior geometric mean of zero for all designs. For such a prior distribution, the Bayesian D-optimality criterion fa...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:Statistica Sinica. - Institute of Statistical Science, Academia Sinica, 1991. - 28(2018), 1, Seite 505-525
1. Verfasser: Waite, Timothy W. (VerfasserIn)
Format: Online-Aufsatz
Sprache:English
Veröffentlicht: 2018
Zugriff auf das übergeordnete Werk:Statistica Sinica
Schlagworte:Philosophy Information science Arts Applied sciences Physical sciences Mathematics
LEADER 01000caa a22002652c 4500
001 JST140205136
003 DE-627
005 20240626021540.0
007 cr uuu---uuuuu
008 240126s2018 xx |||||o 00| ||eng c
035 |a (DE-627)JST140205136 
035 |a (JST)26384251 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Waite, Timothy W.  |e verfasserin  |4 aut 
245 1 0 |a SINGULAR PRIOR DISTRIBUTIONS AND ILL-CONDITIONING IN BAYESIAN D-OPTIMAL DESIGN FOR SEVERAL NONLINEAR MODELS 
264 1 |c 2018 
336 |a Text  |b txt  |2 rdacontent 
337 |a Computermedien  |b c  |2 rdamedia 
338 |a Online-Ressource  |b cr  |2 rdacarrier 
520 |a For Bayesian D-optimal design, we define a singular prior distribution for the model parameters as a prior distribution such that the determinant of the Fisher information matrix has a prior geometric mean of zero for all designs. For such a prior distribution, the Bayesian D-optimality criterion fails to select a design. For the exponential decay model, we characterize singularity of the prior distribution in terms of the expectations of a few elementary transformations of the parameter. For a compartmental model and several multi-parameter generalized linear models, we establish sufficient conditions for singularity of a prior distribution. For the generalized linear models we also obtain sufficient conditions for non-singularity. In the existing literature, weakly informative prior distributions are commonly recommended as a default choice for inference in logistic regression. Here it is shown that some of the recommended prior distributions are singular, and hence should not be used for Bayesian D-optimal design. Additionally, methods are developed to derive and assess Bayesian D-efficient designs when numerical evaluation of the objective function fails due to ill-conditioning, as often occurs for heavy-tailed prior distributions. These numerical methods are illustrated for logistic regression. 
540 |a © 2018 STATISTICA SINICA 
650 4 |a Philosophy  |x Applied philosophy  |x Philosophy of science  |x Scientific method  |x Experimentation  |x Experiment design 
650 4 |a Information science  |x Information analysis  |x Data analysis  |x Regression analysis  |x Logistic regression 
650 4 |a Arts  |x Applied arts  |x Design  |x Design engineering  |x Design efficiency 
650 4 |a Information science  |x Information analysis  |x Data analysis  |x Regression analysis  |x Linear regression  |x Linear models  |x Generalized linear model 
650 4 |a Philosophy  |x Applied philosophy  |x Philosophy of science  |x Scientific method  |x Experimentation  |x Experiment design  |x Factorial design 
650 4 |a Applied sciences  |x Computer science  |x Computer programming  |x Mathematical programming  |x Nonlinear programming  |x Objective functions 
650 4 |a Information science  |x Information analysis  |x Data analysis  |x Regression analysis 
650 4 |a Physical sciences  |x Physics  |x Mathematical physics  |x Numerical quadratures 
650 4 |a Mathematics  |x Mathematical procedures  |x Approximation 
650 4 |a Mathematics  |x Pure mathematics  |x Linear algebra  |x Vector analysis  |x Mathematical vectors 
655 4 |a research-article 
773 0 8 |i Enthalten in  |t Statistica Sinica  |d Institute of Statistical Science, Academia Sinica, 1991  |g 28(2018), 1, Seite 505-525  |w (DE-627)325573409  |w (DE-600)2037676-5  |x 19968507  |7 nnas 
773 1 8 |g volume:28  |g year:2018  |g number:1  |g pages:505-525 
856 4 0 |u https://www.jstor.org/stable/26384251  |3 Volltext 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_JST 
912 |a GBV_ILN_11 
912 |a GBV_ILN_20 
912 |a GBV_ILN_22 
912 |a GBV_ILN_23 
912 |a GBV_ILN_24 
912 |a GBV_ILN_31 
912 |a GBV_ILN_39 
912 |a GBV_ILN_40 
912 |a GBV_ILN_60 
912 |a GBV_ILN_62 
912 |a GBV_ILN_63 
912 |a GBV_ILN_65 
912 |a GBV_ILN_69 
912 |a GBV_ILN_70 
912 |a GBV_ILN_72 
912 |a GBV_ILN_73 
912 |a GBV_ILN_95 
912 |a GBV_ILN_100 
912 |a GBV_ILN_101 
912 |a GBV_ILN_105 
912 |a GBV_ILN_110 
912 |a GBV_ILN_120 
912 |a GBV_ILN_151 
912 |a GBV_ILN_161 
912 |a GBV_ILN_170 
912 |a GBV_ILN_206 
912 |a GBV_ILN_213 
912 |a GBV_ILN_230 
912 |a GBV_ILN_285 
912 |a GBV_ILN_293 
912 |a GBV_ILN_370 
912 |a GBV_ILN_374 
912 |a GBV_ILN_602 
912 |a GBV_ILN_702 
912 |a GBV_ILN_2001 
912 |a GBV_ILN_2003 
912 |a GBV_ILN_2005 
912 |a GBV_ILN_2006 
912 |a GBV_ILN_2007 
912 |a GBV_ILN_2008 
912 |a GBV_ILN_2009 
912 |a GBV_ILN_2010 
912 |a GBV_ILN_2011 
912 |a GBV_ILN_2014 
912 |a GBV_ILN_2015 
912 |a GBV_ILN_2018 
912 |a GBV_ILN_2020 
912 |a GBV_ILN_2021 
912 |a GBV_ILN_2026 
912 |a GBV_ILN_2044 
912 |a GBV_ILN_2050 
912 |a GBV_ILN_2056 
912 |a GBV_ILN_2088 
912 |a GBV_ILN_2107 
912 |a GBV_ILN_2110 
912 |a GBV_ILN_2190 
912 |a GBV_ILN_2949 
912 |a GBV_ILN_2950 
912 |a GBV_ILN_4012 
912 |a GBV_ILN_4027 
912 |a GBV_ILN_4028 
912 |a GBV_ILN_4035 
912 |a GBV_ILN_4037 
912 |a GBV_ILN_4046 
912 |a GBV_ILN_4112 
912 |a GBV_ILN_4116 
912 |a GBV_ILN_4125 
912 |a GBV_ILN_4126 
912 |a GBV_ILN_4155 
912 |a GBV_ILN_4242 
912 |a GBV_ILN_4249 
912 |a GBV_ILN_4266 
912 |a GBV_ILN_4305 
912 |a GBV_ILN_4306 
912 |a GBV_ILN_4307 
912 |a GBV_ILN_4309 
912 |a GBV_ILN_4310 
912 |a GBV_ILN_4311 
912 |a GBV_ILN_4313 
912 |a GBV_ILN_4314 
912 |a GBV_ILN_4315 
912 |a GBV_ILN_4316 
912 |a GBV_ILN_4317 
912 |a GBV_ILN_4318 
912 |a GBV_ILN_4319 
912 |a GBV_ILN_4322 
912 |a GBV_ILN_4323 
912 |a GBV_ILN_4324 
912 |a GBV_ILN_4325 
912 |a GBV_ILN_4326 
912 |a GBV_ILN_4335 
912 |a GBV_ILN_4338 
912 |a GBV_ILN_4346 
912 |a GBV_ILN_4367 
912 |a GBV_ILN_4393 
912 |a GBV_ILN_4700 
951 |a AR 
952 |d 28  |j 2018  |e 1  |h 505-525