Mantovan / Secchi Complex Data Modeling and Computationally Intensive Statistical Methods

1. Auflage 2011
ISBN: 978-88-470-1386-5
Verlag: Springer Milan
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 170 Seiten

Reihe: Contributions to Statistics

Selected from the conference 'S.Co.2009: Complex Data Modeling and Computationally Intensive Methods for Estimation and Prediction,' these 20 papers cover the latest in statistical methods and computational techniques for complex and high dimensional datasets.

Pietro Mantovan has been Professor of Statistics since 1986 at the University Ca' Foscari of Venezia, Italy, where he has served as coordinator of research units, head of the Departement of Statistics, and Dean of the Faculty of Economics. He has written several articles, monographs and textbooks on classical and Bayesian methods for statistical inference. His recent research interests focus on Bayesian methods for learning and prediction, statistical perturbation models for matrix data, dynamic regression with covariate errors, parallel algorithms for system identification in dynamic models, on line monitoring and forecasting of environmental data, hydrological forecasting uncertainty assessment, and robust inference processes.Piercesare Secchi is Professor of Statistics at MOX since 2005 and Director of the Department of Mathematics at the Politecnico di Milano. He got a Doctorate in Methodological Statistics from the University of Trento in 1992 and a PhD in Statistics from the University of Minnesota in 1995. He has written several papers on stochastic games and on Bayesian nonparametric predictive inference and bootstrap techniques. His present research interests focus on statistical methods for the exploration, classification and analysis of high dimensional data, like functional data or images generated by medical diagnostic devices or by remote sensing. He also works on models for Bayesian inference, in particular those generated by urn schemes, on response adaptive designs of experiments for clinical trials and on biodata mining. He is PI of different projects in applied statistics and coordinator of the Statistical Unit of the Aneurisk project.

Autoren/Hrsg.

Mantovan, Pietro

Secchi, Piercesare

Inhaltsverzeichnis

1;Title Page ;1
2;Copyright Page ;4
3;Preface;5
4;Table of Contents;7
5;List of Contributors;9
6;Space-time texture analysis in thermal infraredimaging for classification of Raynaud’s Phenomenon;11
6.1;1 Introduction;11
6.2;2 TheData;12
6.3;3 Processing thermal high resolution infrared images;13
6.3.1;3.1 Segmentation;13
6.3.2;3.2 Registration;13
6.4;4 Feature extraction;15
6.4.1;4.1 ST-GMRFs;16
6.4.2;4.2 Texture statistics through co-occurrence matrices;18
6.5;5 Classification results;19
6.6;6 Conclusions;20
6.7;References;21
7;Mixed-effects modelling of Kevlar fibre failure timesthrough Bayesian non-parametrics;23
7.1;1 Introduction;23
7.2;2 Accelerated life models for Kevlar fibre life data;25
7.3;3 The Bayesian semiparametric AFT model;26
7.4;4 Data analysis;28
7.5;5 Conclusions;34
7.6;Appendix;34
7.7;References;36
8;Space filling and locally optimal designs for Gaussian Universal Kriging;37
8.1;1 Introduction;37
8.2;2 Kriging methodology;39
8.3;3 Optimality of space filling designs;40
8.4;4 Locally optimal designs for Universal Kriging;41
8.4.1;4.1 Optimal designs for estimation;41
8.4.2;4.2 Optimal designs for prediction;46
8.5;5 Conclusions;48
8.6;References;48
9;Exploitation, integration and statistical analysis of thePublic Health Database and STEMI Archive in theLombardia region;50
9.1;1 Introduction;50
9.2;2 The MOMI2 study;52
9.3;3 The STEMI Archive;55
9.4;4 The Public Health Database;56
9.4.1;4.1 Healthcare databases;57
9.4.2;4.2 Health information systems in Lombardia;58
9.5;5 The statistical perspective;58
9.5.1;5.1 Frailty models;59
9.5.2;5.2 Generalised linear mixed models;60
9.5.3;5.3 Bayesian hierarchical models;61
9.6;6 Conclusions;62
9.7;References;62
10;Bootstrap algorithms for variance estimation in PS sampling;65
10.1;1 Introduction;65
10.2;2 The naïve boostrap;66
10.3;3 Holmberg’s PS bootstrap;67
10.4;4 The 0.5 PS-bootstrap;70
10.5;5 The x-balanced PS-bootstrap;70
10.6;6 Simulation study;71
10.7;7 Conclusions;76
10.8;References;76
11;Fast Bayesian functional data analysis of basal body temperature;78
11.1;1 Introduction;78
11.2;2 Methods;80
11.2.1;2.1 RVM in linear models;80
11.2.2;2.2 Extension to linear mixed model;81
11.3;3 Results: application to bbt data;84
11.3.1;3.1 Subject-specific profiles;85
11.3.2;3.2 Subject-specific and population average profiles;86
11.3.3;3.3 Prediction;88
11.4;4 Conclusions;88
11.5;References;89
12;A parametric Markov chain to model age- and state-dependent wear processes ;91
12.1;1 Introduction;91
12.2;2 System description and preliminary technological considerations;93
12.3;3 Data description and preliminary statistical considerations;94
12.4;4 Model description;97
12.5;5 Parameter estimation;99
12.6;6 Testing dependence on time and/or state;101
12.7;7 Conclusions;102
12.8;References;103
13;Case studies in Bayesian computation using INLA;104
13.1;1 Introduction;104
13.2;2 Latent Gaussian models;105
13.3;3 Integrated Nested Laplace Approximation;107
13.4;4 The INLA package for R;108
13.5;5 Case studies;108
13.5.1;5.1 A GLMM with over-dispersion;108
13.5.2;5.2 Childhood under nutrition in Zambia: spatial analysis;110
13.5.3;5.3 A simple example of survival data analysis;115
13.6;6 Conclusions;117
13.7;References;118
14;A graphical models approach for comparing gene sets;120
14.1;1 Introduction;104
14.2;2 Latent Gaussian models;105
14.3;3 Integrated Nested Laplace Approximation;107
14.4;4 The INLA package for R;108
14.5;5 Case studies;108
14.5.1;5.1 A GLMM with over-dispersion;108
14.5.2;5.2 Childhood undernutrition in Zambia: spatial analysis;110
14.5.3;5.3 A simple example of survival data analysis;115
14.6;6 Conclusions;117
14.7;References;118
15;A graphical models approach for comparing gene sets;120
15.1;1 Introduction;120
15.2;2 A brief introduction to pathways;121
15.3;3 Data and graphical models setup;123
15.4;4 Test of equality of two concentration matrices;125
15.5;5 Conclusions;126
15.6;References;126
16;Predictive densities and prediction limits based onpredictive likelihoods;128
16.1;1 Introduction;128
16.2;2 Review on predictive methods;129
16.3;2.1 Plug-in predictive procedures and improvements;130
16.4;2.2 Profile predictive likelihood and modifications;131
16.5;3 Likelihood-based predictive distributions and prediction limits;132
16.5.1;3.1 Probability distributions from predictive likelihoods;133
16.5.2;3.2 Prediction limits and coverage probabilities;135
16.5.3;4 Examples;135
16.5.3.1;4.1 Prediction limits for the sum of future Gaussian observations;136
16.5.3.2;4.2 Prediction limits for the maximum of future Gaussian observations;138
16.5.4;Appendix;139
16.5.5;References;141
17;Computer-intensive conditional inference;142
17.1;1 Introduction;142
17.2;2 An inference problem;144
17.3;3 Exponential family and ancillary statistic models;145
17.4;4 Analytic approximations;146
17.5;5 Bootstrap approximations;147
17.6;6 Examples;149
17.6.1;6.1 Inverse Gaussian distribution;149
17.6.2;6.2 Log-normal mean;150
17.6.3;6.3 Weibull distribution;151
17.6.4;6.4 Exponential regression;152
17.6.5;7 Conclusions;153
17.6.6;References;154
18;Monte Carlo simulation methods for reliability estimation and failure prognostics ;156
18.1;1 Introduction;157
18.2;2 The subset and line sampling methods for realiability estimation;158
18.3;3 Particle filtering for failure prognosis;161
18.4;4 Conclusions;166
18.5;References;167