Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 03 August 2020

(p. 875) Index

(p. 875) Index

The index entries appear in letter-by-letter alphabetical order. Index entries displayed in bold text indicate entries from ‘A’ Appendices.

ability estimation 649
adaptive sampling 826–8, 832–3
adjusted expectation 268–9
Air Pollutants Exposure Model (APEX) 485
air pollution assessment see environmental exposure assessment
alcohol dehydrogenase 29, 36–8
aleatory uncertainty 73–4, 86–7, 408
amino acid sequences see protein bioinformatics
Andersen, T. 604–5
APEX (Air Pollutants Exposure) Model 485
approximate simulation algorithms 166–7
ARF 157, 180
assimilation, sequential 487;
assimilation-contrast effect 601
asynchronous parallel pattern search (APPS) 831, 834–5, 837–8
treed Gaussian process (TGP-APPS) 835, 837–8
ATM kinase 156–7, 158, 180
audio and music processing applications 712–13
computational context and background 742–5
conclusions 742
basic time-domain note and chord models 721–8
Bayesian approach’s advantages 711–12, 742
fundamental tasks 712–13, 717–20
Gamma chains and fields 734–5
Gaussian frequency-domain model 729–30, 731–2
inference in 717–20, 731–3, 742
introduction xxvi, 711–21
latent variance/intensity factorization models 733–4, 735–8
musical transient analysis example 726–8
non-negative matrix factorization (NMF) models 738–41, 744–5
point process frequency-domain model 730, 732–3
polyphonic music transcription 713, 718–20
polyphonic pitch estimation example 739–41
prior structures 722–6, 733–4
properties of musical audio signals 713–16
source separation 713, 718, 743
superposition 715, 716–17
time-domain state-space models 728
tone complex analysis example 731–3
auditing
concepts and terminology 654–5
previous work review 656–7
autoregressive (AR) models 848–52, 864–5, 866–7, 867–8
approximate posterior inference for multi-AR models 856–61, 865, 867–73
moving average (ARMA) model 112
multi-AR model details 851–6, 865
time-varying (TVAR) models 850, 866–7, 867–8
auxiliary feedwater system (AF) 225–6, 230–1, 233
Bayes factors 331–3, 334, 336–7
Bayesian Factor Regression Modelling (BFRM) software 123, 127, 128, 148, 150–1
Bayesian analysis applicational diversity xv
scientific philosophy changes xv
Bayes inference see variational/Bayes inference
(p. 876) Bayes’ theorem, audio-processing context 712
Bernoulli probability 442, 468
BFRM (Bayesian Factor Regression Modelling) software 123, 127, 128, 148, 150–1
BIC (Bayesian information criterion) 10
Bilog-mg 639–40
binary bet 346–7
biodiversity, demographic rates and see demographic rate analysis (tree study)
biomolecule matching and alignment
advantages of Bayesian modelling approach 40
data analysis 36–40
data sources 28–9, 47–8
EM approach 41
future directions 42–3
geometrical transformations 29–30
introduction xvii, 27–30
methodology refinements 40–1
model formulation and inference 44–6
model implementation 46–7
multiconfiguration alignment model 33–6
pairwise matching model 30–3, 44–5
Procrustes type approaches 41–2, 44
protein data and alignment problems 27–9
shape analysis background 43–4
bipolar junction transistors (bjt) study see circuit device experiment design and analysis
bivariate infinite mixtures of experts models 11–12
Black–Scholes model 338
BLAST (Basic Local Alignment Search Tool) 48
breast cancer, and oncogene pathway deregulation see oncogene pathway deregulation and breast cancer
Brownian motion 339
calibration
climate models 410
concept of 752
environmental models 483–4
GP-based 770–2
history matching and 257–63
treed Gaussian processes 823, 833–8
Calvo pricing 374–5, 376, 378, 385, 388
carbonyl reductase 28–9, 36, 37
cardiovascular thrombotic adverse events, Vioxx and see Vioxx trials
CATH 48
causal inference (cancer survival and treating hospital type analysis)
assignment mechanism 684–5, 689–90
as-treated analyses’ problems 697
causal effect of large versus small treating hospitals 689–705
computation 701–2
exclusion restriction 680, 698–9
introduction xxv–xxvi, 679–80
model details 700–1
monotonicity/no-defier assumption 680, 695–7
per-protocol analyses’ problems 697–8
posterior predictive causal inference 685–6
principal stratification 680, 689, 694–7, 706
Rubin’s causal model (RCM) 680–9, 705–6
science assumptions 687–9
stable unit treatment value assumption (SUTVA) 682–3, 691
change point problem 289
chemical Langevin equation (CLE) 167
chemotherapy development 119
Chinese Restaurant Process representation 146, 784, 808;
Cholesky decomposition 773
Chomsky normal form 780
circuit device experiment design and analysis calibration of computer model 823, 833–8
circuit experiments 824
computations 841–2
experimental design 831–3, 841
introduction xxvii, 823–31
treed Gaussian processes 824–31, 838–9, 839–41
validation of computer model 823–4, 833
classification problems 839
climate change projections’ uncertainty characterization
application of latest model projections 562–72
computational details 583–92
conjugate priors context and background 574–6
current state and future scenarios 545–8
diagnostics context and background 582–3
discussion 572–3
extent of data 548–9
future directions 573
hierarchical models’ context and background 576–7
(p. 877) hierarchy of models 551–8
Markov chain Monte Carlo (MCMC) methods’ context and background 577–83
Metropolis–Hastings algorithm background 580–2
model validation 558–62
overview xxiv
simplified datasets 550–1
climate models application examples see climate change predictions’ uncertainty characterization; meridional overturning circulation (MOC) collapse probability analysis
components of 406
climate scenarios and 408
climate system and 406–7
inference from 408–11
quantity and diversity of 545–7
clinical trials missing data/dropout problems 53, 58–66, 66–7
sequential meta-analysis 53–8, 66
Vioxx trial history xvii–xviii, 51–3
CMAQ (Community Multi-Scale Air Quality) Eta forecast model 483, 484, 489–98
code uncertainty 411
cognitive fatigue EEG characterization approximate posterior inference for multi-AR models 856–61, 865, 867–73
autoregressions (AR models) 848–52, 864-5, 866–7, 867–8
computation details 867–73
data description 847–8
EEG data analysis via multi-AR models 861–4
experimental setting 847–8
future directions 865–6
introduction xxvii, 845–8
multi-AR model details 851–6, 865
time series decompositions 848–51, 867
time-varying AR (TVAR) models 850, 866–7, 867–8
Collaborative Perinatal Project, substudy of 4–6
CoMFA (Comparative Molecular Field Analysis) database xvii, 29
Community Multi-Scale Air Quality (CMAQ) Eta forecast model 483, 484, 489–98
complete data likelihood function 59
conditional autoregressive (CAR) model 90, 91, 97, 101–2
conjugacy, inducement of 22
conjugate priors 574–6
continuous time models
financial modelling context and background 338–9
hidden Markov model (CTHMM) 182
coordinate ascent algorithm 789–92, 814–19
Cosmic Background Explorer (COBE) 749
cosmological parameters’ estimation (Λ-cold dark matter (ΛCDM) model)
combined CMB and large scale structure analysis 765–70
cosmic microwave background (CMB) measurement 749
future directions 770
GP-based calibration context and background 770–2
introduction xxvi–xxvii, 749–52
matter power spectrum analysis 752–65
multivariate output emulation context and background 772–3
simulation design 753–4, 755
simulation model overview 752–3
simulator output emulation 754–63, 764, 767–8, 771–3
statistical formulation 763–5
TT spectrum modelling 765–8
covariance structures 91, 97–8, 100–9
Cox proportional hazard regression model 228
cultural differences, educational assessment and see differential item functioning (DIF) (in educational assessment)
Dali database 48
data augmentation (DA) 701–2, 738
DDE, maternal exposure and pregnancy outcome 5–7, 13–18
DDT 13–14;
see also DDE
decision theory 371
decomposition
AR models 848–51, 867
Cholesky 773
multiscale for economic data analysis 295, 296–7, 298, 313, 314–15
singular value (SVD) 756
de Finetti, B. 344, 686
delayed acceptance scheme 772
demographic models 467–8
current limitations 432
multistage model advantages 432–3
demography rate analysis (tree study)
complexity summarizing 463–7
current demographic model limitations 432
data prediction 461–3, 464
demographic data 433–40
demographic model context and background 467–8
diameter growth and fecundity modelling 444–7
exposed canopy area modelling 447
gender and maturation modelling 441–3
hierarchically structured generalized linear model (GLM) background 468–9
incomplete information challenges 431–2
introduction xxii–xxiii, 431–3
multistage model advantages 432–3
prior distributions 448–52
prior/ posterior comparisons 453–61
seed data and fecundity modelling 443–4
survival probability modelling 448
density regression, pregnancy outcomes 13–18
deterministic fixed point iterations 742–5
differential item functioning (DIF) (educational assessment application)
Bayesian methods in item response theory background 649–50
computation context and background 641–5
conclusions 641
DIF analysis of PISA 2003 633–41
DIF overview 624–5, 627–9
introduction xxv, 624–5
model computation details 633, 645–9
model description 629–33
Programme for International Student Assessment (PISA) overview 624–7, 628, 650–1
Dirichlet models 659, 660
context and background for HDP-PFCG and HDP-PCFG-GR 802–7
HDP-PCFG for grammar refinement (HDP-PCFG-GR) 785–7, 793–4, 797–800
hierarchical Dirichlet process probabilistic context-free grammar (HDP-PCFG) 780–5, 787–96, 801
Dirichlet process (DP) 146–7
dependent 20–1
hierarchical (HDP) 805–6
local 11–12, 21
order-based dependent 11, 21
stick-breaking representation of 10–11, 12, 23
Dirichlet process mixture (DPM) models 802, 803–5
definition and graphical model of 804
dependent (DDP) 20–1
finite mixture model background 802–3
latent factor models 127
nonparametric Bayes mixture models context and background 20–1
of normals 10–11
posterior distribution representations for inference 808–9
stick-breaking distribution 803–4, 808–9
Dirichlet process priors 146–7
discrepancy ratio 255
DNA damage 155–8
DNA profiling xix, 188–9
genetic background 209–10
simple paternity testing 190–3
dropout, non-ignorable 53, 58–66, 66–7
Drudge report 606–7
drug development programmes 53, 58, 66;
see also Vioxx trials
dynamic/state-space models 313
for audio and music processing 728
auditing context and background 676–7
computational details of generic 360–1
for demographic modelling 444–5, 470–1
in financial futures markets’ forecasting 350–1
functional network of generic 351
posterior computation in 617–19
dynamic linear models 109–11, 313–14
class of 111–12
future developments in 115
for futures markets’ forecasting 351–2, 354, 355–7, 361–2, 363
general model building approach 111–12
Markov chain Monte Carlo (MCMC) for 112–14, 115, 618
multi-process model approximate posterior distributions 568
in multiscale spatio-temporal model 297, 304–5, 313–14
in multivariate Poisson-lognormal model 90, 97, 99, 109–15
non-normal 113–14
similarities in forecasting and auditing models 676–7
spatio-temporal 114
(p. 879) dynamic stochastic general equilibrium (DSGE) models
policy-making and 370, 371, 391
sequential Monte Carlo (SMC) methods 369, 391–3
Earth system models of intermediate complexity (EMICs) 407, 411–12
econometrics 366
educational assessment, cultural differences and see differential item functioning (DIF) (educational assessment application)
Educational Testing Service (ETS) 628
Efficients Markets Hypothesis (EMH) 347–9, 351
electro-hydraulic control (EHC) system 226–7, 230–2
elicitation
designing 534–7
direct versus indirect 515, 538–40
for regression 514–20
in sequential multilocation auditing 676
software tool for 518, 520–6, 532–3, 534
Elicitator 518, 520–6, 532–3, 534
EM (expectation-maximization) algorithm
in audio and music processing 738
in biomolecule matching and alignment 41
in flexible Bayesian regression 8
in IRT models 649–50
in natural language processing (NLP) 789, 791, 794–801, 807–8, 809–14
variational methods and 742–3, 794–801, 807–8, 809–14
empirical Bayes estimation 304–5
emulation
linear uncertainty analysis 246, 249–53, 253–7, 260, 261–3
multiscale method 253–7
multivariate computer model output 772–3
re-emulation 260, 261–3
stochastic kinetic model 167–8, 182–4
treed Gaussian processes 829–31
energy balance models 407
ensembles of opportunity 547–8
environmental exposure assessment
algorithmic and pseudo-statistical weather prediction 486–8
background and context 503–4
downscaling fusion modelling 490–508
dry deposition estimation 498, 503
environmental computer models 483–6
Gibbs sampling distributions 504–8
history of pollutant space-time modelling 482–3
introduction xxiii, 482–6
National Atmospheric Deposition Program (NADP) data illustration 495–502
stochastic integration fusion modelling 488–9
upscaling fusion modelling 489–90
epidemiologic data
flexible Bayes regression of see pregnancy outcome study
epistemic uncertainty 73–4, 86–7, 408–9
equilibrium, economic definition of 366–7
Euler discretization 325–6, 327, 338–9
evolutionary factor models search 149
exact simulation algorithms 167
exchangeability, second order 268–9
exotic preferences, in DSGE models 390
expectation, adjusted 268–9
expectation-maximization (EM) algorithm see EM (expectation-maximization) algorithm
Expected A-Posteriori (EAP) 649
Expected Global Optimizer (EGO) algorithm 830
exposure–disease relationships
misinterpretation of 3–4
pregnancy outcomes 4–24
factor modelling 144–5
latent factor models 125–7, 146–7, 148–51
(p. 880) failure rate 222–30
false consensus effect 601
filtering theory 368–9
financial futures markets computations 363
dynamic and dynamic linear models 354, 355–7, 360–2, 363
forecasting 350–2
futures markets 344–8, 358
introduction xxi–xxii, 343
portfolio optimization 349–50, 357, 360
probability axioms derivation 359–60
risk modelling 352–8, 359
speculation 348–50, 358–9
subjective expectations 343–4
variance–covariance graphical model 354–7, 362–3
financial market volatility analysis (2007–2008 credit crisis)
conclusions 337–8
continuous-time model context and background 338–9
Garch model 328–9
introduction xxi, 319–24
option prices’ and returns’ informational content 339–40
option prices and volatility 321–4
particle filtering for sequential learning 329–3
problem context and goals 321
results 333–7
stochastic volatility (SV) model 325–6
stochastic volatility jump (SVJ) model 326–8
finite mixture models 7–10, 22, 802–3
flexible Bayes regression, of epidemiologic data 3–24
food safety risk assessment sensitivity analysis
forecasting
DSGE models and 370
dynamic linear models 767
failure 272, 281–5
financial future markets 350–2,
in oil reservoir simulation uncertainty analysis 260, 262, 263–8
prior information and 676
forward filter backward sampler algorithm (FFBS)
Markov-switching stochastic volatility model 619
multiscale spatio-temporal model 298, 305, 314
singular (SFFBS) 298, 305, 316
spatio-temporal mixture model 99–100, 112–13
state-space models 618
frequency/transform domain models
Gaussian 729–30, 731–2
latent variance/intensity factorization 735–7
non-negative matrix factorization 738–41, 744–5
point process 730–1, 732–3
prior structures 733–5
frequentist analysis 6–7, 15–16
full conditional distributions, derivation of 315
fusion modelling
downscaling approach 490–508
stochastic integration approach 488–9
upscaling approach 489–90
Gabor regression 734
Gamma chains 734–5
Gamma field 734–5
Garch model 320, 328–9, 333–8
gas pipeline reliability 289–91
Gaussian covariance function 757
Gaussian process models
calibration 770–3
computer experiment design and 841
drawbacks of 825
emulation 754–63, 771–3, 825
frequency-domain 729–30, 731–2
spatial 503–4;
treed 824–31
Gelman – Rubin procedure 582–3
general circulation models (GCMs) 407
bivariate 556–8
computational details 583–92
future possibilities 572–3
Hadley Center 573
multivariate 554–6
simplified datasets for 550–1
univariate 551–4
validation of 558–62
generalized linear models (GLMs) 113–14, 468–9, 514
generalized method of moments (GMM) 369
(p. 881) GENIE-1 (Grid Enabled Integrated Earth system model) 411–12
Geo-R 504
geostatical modelling 90–1
Gibbs sampling 643
for climate models 556, 558, 560–1, 578–80, 583–92
collapsed/marginal/Polya Urn 22–3, 146–7;
conditional/blocked 23–4
for demographic model 452–3, 469–79
for environmental exposure assessment 504–8
for IRT models 650
for latent factor regression models 148
mean field and 815
for multiscale spatio-temporal model 305
for multivariate Poisson-lognormal model 99
non-parametric Bayes analysis 9–10, 22, 22–4
for state-space models 618
for stochastic kinetic model 168
for treed Gaussian processes 842
variational method alternative to 742–3, 745
Gillespie algorithm 161, 166–7
graph theoretic technique 36, 38
graphical models
in audio and music processing 720, 734–5
in natural language processing (NLP) 781, 788
variance–covariance 354–7, 362–3
greenhouse gases emission scenarios (SRES) 414, 546–7, 548–9, 550
Grid Enabled Integrated Earth system model (GENIE-1) 411–12
Hadley Center GCM 573
Hansen, L. 369
Hastings step 765;
Health Maintenance Organizations (HMOs) 675
health outcomes
exposure–disease relationship misinterpretation 3–4, 19
pregnancy study see pregnancy outcome study
Heidelberger – Welch procedures 583–4
hidden Markov models (HMMs) 181, 792, 805
continuous-time (CTHMM) 182
hierarchical Dirichlet process (HDP-HMM) 782, 783, 805–7
probability context-free grammars (PCFGs) and 781
hierarchically-structured generalized linear model (GLM) 468–9
history matching 257–63
17–β hydroxysteroid-dehydrogenase 28–9, 36, 37
hypoxia-related pathways 143
identifiability, model 633
implausibility 257–61
independent components analysis 418
information criterion, Bayesian (BIC) 10
information ratio 349, 352
inside-outside algorithm 781
instrumental variables models 706
intensity factorization models 735–8
Intergovernmental Panel on Climate Change (IPCC) 408, 414, 546
interpolation, spatial see spatial interpolation
Intrade/Tradesports 598, 600–1, 606
in-vitro to in-vivo translation problem 118, 119–20
Iowa Electronic Markets (IEM) 598, 600–2
item response theory (IRT)
Bayesian methods in 649–50
(p. 882) in educational assessment 629–33;
Kalman filter
for DSGE models 369, 378, 381, 394–5
in particle filtering 330
sequential assimilation and 487–8
Kaplan–Meier curves 135–7
kernel continuation ratio probit model 22
kernel stick-breaking process (KSBP) 11–13, 14, 15, 21
KL-divergence 814–15
KL-projection 794
Kronecker structure 773
Kolmogorov–Smirnov statistics 292
Kydland, F. 367–8
label ambiguity 8–9
Lagrangian function 373
latent class model 7–10
latent factor models 125–7
Dirichlet process priors in 146–7
models and computations 148–51
latent mean process reconstruction 305–6
latent point process model 30–1
latent variance models 733–4, 735–7
latin hypercube sampling climate models 412–13
cosmological models 754, 755, 761–2, 763
stochastic kinetic model 183
treed Gaussian processes 835–7
Legacy Archive for Microwave Background Data Analysis (LAMBDA) 767
legal evidence issues 208–9
likelihood
joint 210–13
in legal evidence 208–9
linear methods 263–8, 268–9
local partition process 21
logistic regression, elicitation for 512, 514–20
Lucas, R. 367–8
Lucas’ critique 368
macroeconomics 366
main effect 80–1, 82–3
main effect index 80, 85–6
maintenance model 222–3
malaria
DDT use 13–14
disease mapping see malaria mapping
prevalence and impact of 92
malaria mapping (spatio-temporal mixture model)
dynamic linear model (DLM) background and context 109–15
introduction xviii, 90–1
model extension possibilities 109
motivation for 92–6
multivariate Poisson-lognormal model 96–102
results 102–9
Markov chain Monte Carlo (MCMC) methods 577–80, 641–5
in biomolecule matching and alignment 37–40, 41, 45–7
block-move 618
for causal inference 701–2
for climate models 577–83, 583–92
in clinical trials’ analysis 62, 64, 65
data augmentation (DA) 701–2
for DIF models 645–9
in disease mapping 97, 99, 100–1, 102
for dynamic linear models (DLMs) 112–14, 115, 314, 618
for dynamic stochastic general equilibrium (DSGE) models 369, 370
for flexible Bayesian regression 8–9, 13, 15
for latent factor regression models 127, 148
for Markov-switching stochastic volatility model 606, 617–19
for multiscale spatio-temporal model 305–6
in reliability analysis 279–81
for semiparametric model 229
single-move 618
in sparse multivariate regression analysis 123
for stochastic kinetic model 155, 168–70, 171, 174, 181, 182
for stochastic volatility models 326
for treed Gaussian processes 827, 829
variational alternatives and 742–5, 807–8, 809–14
Markovian process models
classification 181
Markov-switching stochastic volatility model 604–6;
parameter inference and time course data 181–2
Markov random field (MRF) 734
mean field 787–9, 742–3
for DP-based models 809–14
update equation derivation 814–19
measurement error model (MEM) 490
meridional overturning circulation (MOC) collapse probability analysis
21st century MOC simulations 415–17
climate models 406–8
climate simulator uncertainty 408–9
context and background 425–7
data and climate prediction 409–10
experimental structure 412–14
future directions 424–5
GENIE-1 411–12
input parameter bounds’ elicitation 414–15
introduction xxii, 403–6
Monte Carlo methods 410
results summary 424
uncertainty analysis 419–24
method-of-moments analysis 696, 697, 698
generalized 369
Metropolis step 453, 470–9, 759, 765;
Metropolis – Hastings step 47, 229, 580–2, 643–5
biomolecule matching and alignment 47
climate models 554, 556, 558, 580–2, 583–92
DIF models 645–8
DSGE model 382, 394
mixture of Polya trees (MPT) model 229
multivariate Poisson-lognormal model 99, 113
non-normal DLMs 113
state-space models 618
treed Gaussian processes 842
Metropolis-within-Gibbs algorithm 644–5, 650
microbial risk assessment sensitivity analysis (VTEC 0157 study)
contamination assessment model 72–3
epistemic and aleatory uncertainty context and background 86–7
future directions 84
introduction xviii, 69
microbial risk assessment background 69–70
model input distributions 73–9
model output analysis 79–84
problem of Vero-cytotoxic E.coli 0157 (VTEC 0157) in pasteurized milk 70–2
variance-based sensitivity background 84–6
missing data problems
in causal inference 681–2
in clinical trials 53, 58–66, 66–7
mixture of Polya trees models (MPTs) 228–31, 233–5
Model A-Posteriori (MAP) 649
model identifiability 633
modern portfolio theory (MPT) 349–50
Molecular Signatures Database (MSigDB) 142
monetary unit sampling 654, 656–7
monotonicity/no-defier assumption 680, 695–7
Monte Carlo (MC) estimators 642
Monte Carlo (MC) methods
for climate models 410, 420
in microbial risk assessment 70, 79–82, 86
for sequential multilocation auditing 662
for treed Gaussian processes 829
multi-AR models 851–6, 865
approximate posterior inference for 856–61, 865, 867–73
multiscale modelling
background 313
multiscale spatio-temporal model (agricultural production analysis)
agricultural production data analysis 300–2, 306–11, 312
computing background 315–16
dynamic linear model context and background 313–14
dynamic multiscale modelling 302–4
estimation 304–6
introduction xxi, 295–8
massive data set capability 311, 313
model summary 304
multiscale decomposition 295, 296–7, 298, 313, 314–15
multiscale factorization 298–300
multiscale model background 313
music processing see audio and music processing
mutation models 194–6
mutation rates, uncertainty and paternity testing see paternity testing and mutation rate uncertainty
MySQL 522–3
(p. 884) National Ambient Air Quality Standards (NAAQS) Exposure Model 485
national and oncome product accounts (NIPA) 396
National Atmospheric Deposition Program (NADP) 495–502
natural language processing (syntactic parsing problem)
adaptor grammar framework 784–5
approximate posterior inference algorithm for HDP-PCFG 787–94, 801, 807–8, 814–19
Bayesian approach’s advantages 776
DP-based models’ context and background 802–7
experiments 794–800
future directions 801
grammar induction 777–8, 794–6
grammar refinement 778–9, 785–7, 793–4, 797–801
HDP-PCFG for grammar refinement (HDP-PCFG-GR) 785–7, 793–4, 797–800
hidden Markov models (HMMs) and 777
hierarchical Dirichlet process probabilistic context-free grammar (HDP-PCFG) model 780–5, 787–96, 801
infinite tree framework 784, 785
introduction xxvii, 776–81
latent-variable models 779
lexicalization 778–9
machine translation 801
multilingual data modelling 801
parent annotation 778, 779
parse trees and treebanks 776–7, 778–9, 789–91, 792–4, 798–800
Penn treebank parsing 798–800
probabilistic context-free grammars (PCFGs) 777–8, 780–81
update equations’ derivations 814–19
variational inference context and background 807–14
net payoff 349
New Keynesian DSGE models 369, 371–2
aggregation 377–8
equilibrium 378–80
final good producer 375
future directions 390–1
the government 376–7
households 372–5
intermediate good producers 375–6
likelihood function 380–2
US economy data estimation 382–90
new macroeconometric US economy dynamics analysis (DSGE model)
data construction 396
empirical analysis 382–90
future directions 390–1
introduction xxii, 366–71
Kalman filter 369, 378, 381, 394–5
model computation 393–4
new Keynesian model 371–82
policy-making 370, 371, 391
sequential Monte Carlo (SMC) methods 369, 391–3
Newtonian relaxation 487
NIPA (national and oncome product accounts) 396
no-defier/monotonicity assumption 680, 695–7
non-negative matrix factorization (NMF) models 738–41, 744–5
nonparametric models 10–12, 20–2
Normal-Gamma model 759, 764
NPACI/NBCR resource 48
nuclear power plant maintenance analysis and decisions
data 224–7
introduction xx, 219–22
maintenance model 222–3
mixture of Polya trees models (MPTs) 228–31, 233–5
optimization algorithm 238
optimization results 223–4
parametric model 224–5, 227–8, 230–3
proof of propositions 235–9
semiparametric model 225, 228–31
nudging 487
oestrogen receptor (ER) pathways 143–4
Office of Air Quality and Planning 485
oil reservoir simulation uncertainty analysis
analysis overview 246–7
Bayes linear approach context and background 268–9
coarse model emulation 249–53
forecasting 263–8
general formulation of approach 244–6
history matching and calibration 257–63
initial model runs and screening 247–9
introduction xx, 241–2
model description 242–4
multiscale emulation 253–7
oncogene pathway deregulation and breast cancer
(p. 885) BFRM (Bayesian Factor Regression Modelling) software 123, 127, 128, 148, 150–1
clinical outcome prediction 133–40
Dirichlet process priors context and background 146–7
evaluation and pathway annotation analysis 140–4, 150–1
evolutionary factor models search 149
factor models’ context and background 144–5
introduction xviii–xix, 118
latent factor models for tumour gene expression 125–7, 144–7, 148–51
latent factor projection from in-vivo to in-vitro 131–3, 134
latent factor structure exploration 128–31
MCMC for 148, 149, 151
pathway activity quantification 123–5
pathway signature generation 122–3
previous studies 120–1
probabilistic pathway annotation (PROPA) analysis 142–4, 150–1
problem context and goals 118–20
shotgun stochastic search (SSS) 133–5, 150
sparsity modelling in multivariate analysis background 147
Weibull survival regression models 133–40, 150
opinion polls 598
optimization 787, 789
asynchronous parallel pattern search (APPS) 831, 834–5, 837–8
climate model 427
Expected Global Optimizer algorithm 830
global sensitivity analysis and 828
maintenance analysis 219, 221, 222–4, 238
Pareto 349, 357
portfolio 349–50, 357, 360
treed Gaussian processes 829–31, 834–5, 837–8
option prices 319
informational content 339–40
and volatility 321–4
Organization for Economic Co-Operation and Development (OECD)
Programme for International Student Assessment (PISA) see differential item functioning (DIF) (in educational assessment)
palaeoclimate data, use of 405–6, 409
parameter relationship modelling 658–9
parametric model 224–5, 227–8, 230–3
Pareto optimization 349, 357
particle filtering
for DSGE models 392–3
in volatility analysis 319–21, 329–33, 337
paternity testing and mutation rate uncertainty allele frequency uncertainty 207–8
analytical allowance for mutation 193–7
assumed mutation rate case analysis 197–8
Bayesian networks 207–8
case analysis example 191–2, 197–8, 206
casework data 200–1
DNA profiling context and background 209–10
introduction xix, 188–90
joint likelihood 210–13
legal evidence context and background 208–9
mutation rate data analysis 204–5
mutation rate likelihood 201–4
mutation rate uncertainty 198–200
null alleles 207
simple cases 190–3
unobtainable DNA profiles 207
pCNEM 485
Penn Treebank 798–800
perturbation methods 378–9, 381–2, 394
pharmaceutical development/testing see drug development programmes; Vioxx trials
PISA (Programme for International Student Assessment), cultural differences and see differential item functioning (DIF) (in educational assessment)
plant population demography analysis see demography rate analysis (tree study)
point process frequency-domain model 730, 732–3
Poisson process models 272
Poisson-lognormal model (multivariate) application example see malaria mapping
for repairable systems 286–92
zero-inflated Poisson (ZIPo) model 469
policy-making, DSGE models and 370, 371, 391
pollution exposure assessment see environmental exposure assessment
Polya trees 233–5
mixtures of (MPT) 228–31, 233–5
Polya urn representation 22–3, 146–7;
portfolio optimization 349–50, 360
power law process (PLP) 278, 288–9, 293
prediction market volatility (political campaign information flow measurement)
computation context and background 617–20
conclusions 615–16
(p. 886) future directions 617
introduction xxiv, 597–8
model description 604–6
political prediction markets 599–602
US presidential election 2004 xxiv, 597–9, 606–16
volatility, trading volume and information flow 602–8
pregnancy outcome study
density regression 33–8
finite mixture models 7–10
flexible Bayesian regression 7–24
introduction xvii, 3–7
Kernel stick-breaking process 12–13
nonparametric Bayes methods 10–12
nonparametric Bayes mixure models’ context and background 20–2
posterior computation 22–3
standard regression approach’s limitations 4–7, 19
Prescott, E. 367–8
principal components analysis 418
principal stratification 680, 689, 694–7, 706
principal variables 247–9
prior information 676
probabilistic context-free grammars (PCFGs) 777–8, 780–81
hierarchical Dirichlet processes and see natural language processing (syntactic parsing problem)
probabilistic pathway annotation (PROPA) analysis 142–4, 150–1
probability, Bayes definition of 346
probability axioms, derivation of 359–60
probability theory, origins of 343
probit stick-breaking process (PSBP) 22
Procrustes analysis 41–2, 44
Program for Climate Model Diagnosis and Intercomparison (PCMDI) 547, 548–9, 562
Programme for International Student Assessment (PISA), cultural differences and see differential item functioning (DIF) (in educational assessment)
propositions, proof of 235–8
protein bioinformatics study active sites 28–9
advantages of Bayesian approach 40
alternative approaches 41–2
data analysis 36–40
data sources 28–9, 47–8
future directions 42–3
introduction to xvii, 27–30
methodology refinements 40–1
model formulation and inference 44–6
model implementation 46–7
multiconfiguration alignment model 33–6
pairwise matching model 30–3
shape analysis context and background 43–4
PSB databank 47–8
quantile regression 5
random effect terms 90, 96, 109, 503
Rayleigh quotient 353
regression
density 33–8
dynamic linear models 112
elicitation for 514–20
logistic 3–7, 8, 19
sparse multivariate application example see oncogene pathway deregulation and breast cancer
reliability analysis
gas pipelines 289–91
repairable systems
Poisson process models for 286–92
reliability analysis see train door reliability analysis
replication 682
resample-propagate filter 330–1
response and missingness, joint model 59–61
reversible jump Markov chain Monte Carlo (RJMCMC)
audio and music processing 722, 725–8, 731–3
finite mixture models 10
Poisson process models 289
treed Gaussian processes and 826, 842
risk modelling, financial futures markets xxi, 352–8, 359
rock-wallaby habitat modelling and mapping (indirect elicitation from experts)
elicitation design 523, 534–7
elicitation method 514–23, 538–40
encoding 523–7, 537
introduction xxiii–xxiv, 511–12
opinion collation 527–31
outcome discussion 531–4
problem introduction 511–14
root mean square deviations (RMSD) 32, 44
Rubin’s causal model (RCM)
application example see causal inference (cancer survival and treating hospital type analysis)
framework 680–9
observational studies background 705–6
(p. 887) sampling, adaptive 826–8, 832–3
Savage–Dickey density ratio 332
scientific philosophy, Bayesian analysis and changes in xv
SCOP 48
seasonal models 112
selection model 60
semiparametric model 225, 228–31
sensitivity analysis microbial risk assessment see microbial risk assessment sensitivity analysis (VTEC 0157 study)
paternity testing 197–8
sequential multilocation auditing 671–4, 675
treed Gaussian processes 828–9, 835, 836
of univariate simulator 759–60
variance-based 80–1, 82, 84–6, 828–9
Sequential Design and Analysis of Computer Experiments (SDACE) 841
Sequential Design of Experiments (DOE) 841
sequential estimation 329–33, 336
sequential meta-analysis 53–8, 66
sequential Monte Carlo (SMC) methods for DSGE models 369, 391–3
for Markov process models 181, 182
for multi-AR models 856, 865
sequential multilocation auditing (New York food stamps program application)
auditing concepts and terminology 654–5
dynamic model context and background 676–7
error classes modelling 660–1
error rates modelling 659–60
introduction xxv, 653–8
New York food stamps program 653–4, 655, 666–75
notation and outline 657–8
previous work review 656–7
prior information context and background 676
projection 662–5
sensitivity analysis 671–4, 675
updating 661–2
wider applicability 675–6
serum fibroblast cell cycle pathway gene list 142–3
shape analysis xvii, 28, 43–4;
short tandem repeat (STR) markers 193, 210
shotgun stochastic search (SSS) 133–5, 150
signal-to-noise ratio 303, 306–9
similarity index 31
singular forward filter backward sample (SFFBS) 298, 305, 316
SitesBase 48
Sloan Digital Sky Survey (SDSS) 749–52
South Texas Project (STP) Electric Generation Station 219–20
sparse multivariate regression analysis 122–3, 124, 147
sparsity priors 123, 147
spatial correlation 90–1, 107–9
spatial data Bayesian software 504
spatial Gaussian process models 503–4
application example see environmental exposure assessment
spatial interpolation environmental exposure assessment 494–5, 496, 498–9, 501
malaria mapping 100–1, 104, 106
spatial random effects 503
spatio-temporal models
application example see malaria mapping
dynamic linear 114
SpBayes 504
Special Report on Emissions Scenarios (SRES) 414, 547, 548–9, 550
speculation 348–50, 358–9
stable unit treatment value assumption (SUTVA) 682–3, 691
standardization, internal 94, 109
standardized mortality ratio (SMR) 94–6
state-space models see dynamic/state-space models
steady model 111
steroid molecules, matching multiple configurations of 29, 38–40
Stochastic Human Exposure and Dose Simulation (SHEDS) model 485
stochastic kinetic model of p53 and Mdm2 oscillations
data 161–2
emulator construction 167, 182–4
inference from multiple cells 173–80
inference from single cell data 170–3, 174
introduction xix, 155–60
linking model to data 162–6
Markov process models’ context and background 181–2
model 160–1
model extension possibilities 180–1
parameter inference difficulties 181–2
posterior computation 166–70
(p. 888) stochastic singularity problem 380
stochastic volatility (SV) model 320, 324, 325–6, 331, 332–8, 339
in audio and music processing 734
stochastic volatility with jumps (SVJ) model 320, 324, 326–8, 331, 332–8
STP Nuclear Operating Company (STPNOC) 219–20
successive corrections method (SCM) 486–7
systems biology application example see stochastic kinetic model of p53 and Mdm2 oscillations
Systems Biology Markup Language (SBML) 158–9
Tamoxifen 144
Taylor series representation 111
temporal prediction 101, 103–4, 105
temporal sure preference principle 269
three-parameter logistic model 630–3
time domain audio models 721–8
Time to Build and Aggregate Fluctuations 367–8
Tor Fjeldy photocurrent model 824
total effect variances 85–6
Total Risk Integrated Methodology (TRIM) model framework 485
Tradesports/Intrade 598, 600–1, 606
train door reliability analysis (bivariate Poisson process model)
data quality problems 272, 285–6
exploratory data analysis 273–6
failure forecasting 272, 281–5
intensity function 275–8, 281, 285, 287–92
introduction xx–xxi, 271–3
likelihood, prior and parameter estimation 278–81
model development possibilities 285
Poisson process models for repairable systems 286–92, 293
warranty parameter assessment 272–3, 281–5
transform analysis 339–40
transform domain models see frequency/transform domain models
translation problem, in-vitro to in-vivo 118, 119–20
treatment effect 62–6
tree demography rate analysis see demography rate analysis (tree study)
treed Gaussian processes (TGP) 824–6, 839–41
adaptive sampling 826–8
future directions 839
optimization 829–31, 834–5, 837–8
potential applications 838–9
sensitivity analysis 828–9, 835, 836
uncertainty, aleatory and epistemic 73–4, 86–7, 408
uncertainty analysis
US Environmental Protection Agency (EPA) 485
US presidential election (2004), information flow measurement see prediction market volatility (political campaign information flow measurement)
variational/ Bayes inference 742–5
approximate posterior inference algorithm for HDP-PCFG 787–801, 807–14, 814–19
AR and TVAR models 867–8
comparison with Markov chain Monte Carlo sampling 807–8
DP mixture models representation and 808–9
mean-field inference and DP-based models 809–12
and prior interaction 812–14
Vero-cytotoxigenic E. coli (VTEC) 0157 study see microbial risk assessment sensitivity analysis (VTEC 0157 study)
Vioxx trials study
APPROVe study 52, 54–5, 58–66
background to xvii–xviii, 51
(p. 889) cardiovascular adverse events’ definition 54
non-ignorable dropout 58–66, 66–7
placebo-controlled trials list 52
sequential meta-analysis 53–8, 66
VIGOR trial 55, 56
Viterbi algorithm 793
VIX index 319, 321, 322–5
VXN index 319, 321, 322–5
Wall Street Journal, Penn Treebank 798–800
warranty parameter assessment 272–3, 281–5
wavelet decomposition 296, 298, 313
weather prediction climate forecasting and 559
numerical approaches 486–8
Weibull survival models 133–40, 150
Wilkinson Microwave Anisotropy Probe (WMAP) 749, 765–6, 768–80
WinBUGS 504, 522
wish fulfillment 601, 602
Xyce 824, 834
zero-inflated Poisson (ZIPo) model 469