Value of information in the earth sciences :
Eidsvik, Jo.
Value of information in the earth sciences : integrating spatial modeling and decision analysis / Jo Eidsvik, Tapan Mukerji, and Debarun Bhattacharjya. - xiii, 385 pages : illustrations (black and white) ; 26 cm
Includes bibliographical references and index.
Machine generated contents note: 1.Introduction
1.1.What is the value of information?
1.2.Motivating examples from the Earth sciences
1.3.Contributions of this book
1.4.Organization
1.5.Intended audience and prerequisites
1.6.Bibliographic notes
2.Statistical models and methods
2.1.Uncertainty quantification, information gathering, and data examples
2.2.Notation and probability models
2.2.1.Univariate probability distributions
2.2.2.Multivariate probability distributions
2.3.Conditional probability, graphical models, and Bayes' rule
2.3.1.Conditional probability
2.3.2.Graphical models
2.3.3.Bayesian updating from data
2.3.4.Examples
2.4.Inference of model parameters
2.4.1.Maximum likelihood estimation
2.4.2.Examples
2.5.Monte Carlo methods and other approximation techniques
2.5.1.Analysis by simulation
2.5.2.Solving integrals
2.5.3.Sampling methods
2.5.4.Example
2.6.Bibliographic notes
Contents note continued: 3.Decision analysis
3.1.Background
3.2.Decision situations: terminology and notation
3.2.1.Decisions, uncertainties, and values
3.2.2.Utilities and certain equivalent
3.2.3.Maximizing expected utility
3.2.4.Examples
3.3.Graphical models
3.3.1.Decision trees
3.3.2.Influence diagrams
3.3.3.Examples
3.4.Value of information
3.4.1.Definition
3.4.2.Perfect versus imperfect information
3.4.3.Relevant, material, and economic information
3.4.4.Examples
3.5.Bibliographic notes
4.Spatial modeling
4.1.Goals of stochastic modeling of spatial processes
4.2.Random fields, variograms, and covariance
4.3.Prediction and simulation
4.3.1.Spatial prediction and Kriging
4.3.2.Common geostatistical stochastic simulation methods
4.4.Gaussian models
4.4.1.The spatial regression model
4.4.2.Optimal spatial prediction: Kriging
4.4.3.Multivariate hierarchical spatial regression model
4.4.4.Examples
Contents note continued: 4.5.Non-Gaussian response models and hierarchical spatial models
4.5.1.Skew-normal models
4.5.2.Spatial generalized linear models
4.5.3.Example
4.6.Categorical spatial models
4.6.1.Indicator random variables
4.6.2.Truncated Gaussian and pluri-Gaussian models
4.6.3.Categorical Markov random field models
4.6.4.Example
4.7.Multiple-point geostatistics
4.7.1.Algorithms
4.7.2.Example
4.8.Bibliographic notes
5.Value of information in spatial decision situations
5.1.Introduction
5.1.1.Spatial decision situations
5.1.2.Information gathering in spatial decision situations
5.1.3.Overview of models
5.2.Value of information: a formulation for static models
5.2.1.Prior value
5.2.2.Posterior value
5.2.3.Special cases: an overview
5.3.Special case: low decision flexibility and decoupled value
5.3.1.Prior value
5.3.2.Posterior value
5.3.3.Computational notes
5.3.4.Example
Contents note continued: 5.4.Special case: high decision flexibility and decoupled value
5.4.1.Prior value
5.4.2.Posterior value
5.4.3.Computational notes
5.4.4.Examples
5.5.Special case: low decision flexibility and coupled value
5.5.1.Prior value
5.5.2.Posterior value
5.5.3.Computational notes
5.5.4.Example
5.6.Special case: high decision flexibility and coupled value
5.6.1.Prior value
5.6.2.Posterior value
5.6.3.Computational notes
5.6.4.Example
5.7.More complex decision situations
5.7.1.Generalized risk preferences
5.7.2.Additional constraints
5.7.3.Sequential decision situations
5.8.Sequential information gathering
5.9.Other information measures
5.9.1.Entropy
5.9.2.Prediction variance
5.9.3.Prediction error
5.10.Bibliographic notes
6.Earth sciences applications
6.1.Workflow
6.2.Exploration of petroleum prospects
6.2.1.Gotta get myself connected: Bayesian network example
Contents note continued: 6.2.2.Basin street blues: basin modeling example
6.2.3.Risky business: petroleum prospect risking example
6.3.Reservoir characterization from geophysical data
6.3.1.Black gold in a white plight: reservior characterization example
6.3.2.Reservoir dogs: seismic and electromagnetic data example
6.4.Mine planning and safety
6.4.1.I love rock and ore: mining oxide grade example
6.4.2.We will rock you: rock hazard example
6.5.Groundwater management
6.5.1.Salt water wells in my eyes: groundwater management example
6.6.Bibliographic notes
7.Problems and projects
7.1.Problem and tutorial hands-on projects
7.1.1.Problem sets
7.1.2.Hands-on projects
7.2.Hands on: exploration of petroleum prospects
7.2.1.Gotta get myself connected: Bayesian network example
7.2.2.Basin street blues: basin modeling example
7.2.3.Risky business: petroleum prospect risking example
Contents note continued: 7.3.Hands on: reservoir characterization from geophysical data
7.3.1.Black gold in a white plight: reservoir characterization example
7.3.2.Reservoir dogs: seismic and electromagnetic data example
7.4.Hands on: mine planning and safety
7.4.1.I love rock and ore: mining oxide grade example
7.4.2.We will rock you: rock hazard example
7.5.Hands on: groundwater management
7.5.1.Part I: salt water wells in my eyes
groundwater monitoring in Netica
7.5.2.Part II: salt water wells in my eyes
groundwater monitoring in BNT
Appendix: selected statistical models and sampling methods
Appendix A.1 Gaussian distribution
A.1.1.Definition and properties
A.1.2.Decision analysis and VOI results
Appendix A.2 Generalized linear models
A.2.1.Definition and properties
A.2.2.Decision analysis and VOI results
Appendix A.3 Markov chains and hidden Markov models
A.3.1.Definition and properties
Contents note continued: A.3.2.Decision analysis and VOI results
Appendix A.4 Categorical Markov random fields
A.4.1.Definition and properties
A.4.2.Decision analysis and VOI results
Appendix A.5 Discrete graphs and Bayesian networks
A.5.1.Definition and properties
A.5.2.Decision analysis and VOI results
Appendix B Sampling methods.
Presents a unified framework for assessing the value of potential data gathering schemes, with a focus on the Earth sciences.
9781107040267 (hbk.)
2015022572
Earth sciences--Information services.
Geochemical prospecting.
Communication in science.
550.01156 EID / 009925
Value of information in the earth sciences : integrating spatial modeling and decision analysis / Jo Eidsvik, Tapan Mukerji, and Debarun Bhattacharjya. - xiii, 385 pages : illustrations (black and white) ; 26 cm
Includes bibliographical references and index.
Machine generated contents note: 1.Introduction
1.1.What is the value of information?
1.2.Motivating examples from the Earth sciences
1.3.Contributions of this book
1.4.Organization
1.5.Intended audience and prerequisites
1.6.Bibliographic notes
2.Statistical models and methods
2.1.Uncertainty quantification, information gathering, and data examples
2.2.Notation and probability models
2.2.1.Univariate probability distributions
2.2.2.Multivariate probability distributions
2.3.Conditional probability, graphical models, and Bayes' rule
2.3.1.Conditional probability
2.3.2.Graphical models
2.3.3.Bayesian updating from data
2.3.4.Examples
2.4.Inference of model parameters
2.4.1.Maximum likelihood estimation
2.4.2.Examples
2.5.Monte Carlo methods and other approximation techniques
2.5.1.Analysis by simulation
2.5.2.Solving integrals
2.5.3.Sampling methods
2.5.4.Example
2.6.Bibliographic notes
Contents note continued: 3.Decision analysis
3.1.Background
3.2.Decision situations: terminology and notation
3.2.1.Decisions, uncertainties, and values
3.2.2.Utilities and certain equivalent
3.2.3.Maximizing expected utility
3.2.4.Examples
3.3.Graphical models
3.3.1.Decision trees
3.3.2.Influence diagrams
3.3.3.Examples
3.4.Value of information
3.4.1.Definition
3.4.2.Perfect versus imperfect information
3.4.3.Relevant, material, and economic information
3.4.4.Examples
3.5.Bibliographic notes
4.Spatial modeling
4.1.Goals of stochastic modeling of spatial processes
4.2.Random fields, variograms, and covariance
4.3.Prediction and simulation
4.3.1.Spatial prediction and Kriging
4.3.2.Common geostatistical stochastic simulation methods
4.4.Gaussian models
4.4.1.The spatial regression model
4.4.2.Optimal spatial prediction: Kriging
4.4.3.Multivariate hierarchical spatial regression model
4.4.4.Examples
Contents note continued: 4.5.Non-Gaussian response models and hierarchical spatial models
4.5.1.Skew-normal models
4.5.2.Spatial generalized linear models
4.5.3.Example
4.6.Categorical spatial models
4.6.1.Indicator random variables
4.6.2.Truncated Gaussian and pluri-Gaussian models
4.6.3.Categorical Markov random field models
4.6.4.Example
4.7.Multiple-point geostatistics
4.7.1.Algorithms
4.7.2.Example
4.8.Bibliographic notes
5.Value of information in spatial decision situations
5.1.Introduction
5.1.1.Spatial decision situations
5.1.2.Information gathering in spatial decision situations
5.1.3.Overview of models
5.2.Value of information: a formulation for static models
5.2.1.Prior value
5.2.2.Posterior value
5.2.3.Special cases: an overview
5.3.Special case: low decision flexibility and decoupled value
5.3.1.Prior value
5.3.2.Posterior value
5.3.3.Computational notes
5.3.4.Example
Contents note continued: 5.4.Special case: high decision flexibility and decoupled value
5.4.1.Prior value
5.4.2.Posterior value
5.4.3.Computational notes
5.4.4.Examples
5.5.Special case: low decision flexibility and coupled value
5.5.1.Prior value
5.5.2.Posterior value
5.5.3.Computational notes
5.5.4.Example
5.6.Special case: high decision flexibility and coupled value
5.6.1.Prior value
5.6.2.Posterior value
5.6.3.Computational notes
5.6.4.Example
5.7.More complex decision situations
5.7.1.Generalized risk preferences
5.7.2.Additional constraints
5.7.3.Sequential decision situations
5.8.Sequential information gathering
5.9.Other information measures
5.9.1.Entropy
5.9.2.Prediction variance
5.9.3.Prediction error
5.10.Bibliographic notes
6.Earth sciences applications
6.1.Workflow
6.2.Exploration of petroleum prospects
6.2.1.Gotta get myself connected: Bayesian network example
Contents note continued: 6.2.2.Basin street blues: basin modeling example
6.2.3.Risky business: petroleum prospect risking example
6.3.Reservoir characterization from geophysical data
6.3.1.Black gold in a white plight: reservior characterization example
6.3.2.Reservoir dogs: seismic and electromagnetic data example
6.4.Mine planning and safety
6.4.1.I love rock and ore: mining oxide grade example
6.4.2.We will rock you: rock hazard example
6.5.Groundwater management
6.5.1.Salt water wells in my eyes: groundwater management example
6.6.Bibliographic notes
7.Problems and projects
7.1.Problem and tutorial hands-on projects
7.1.1.Problem sets
7.1.2.Hands-on projects
7.2.Hands on: exploration of petroleum prospects
7.2.1.Gotta get myself connected: Bayesian network example
7.2.2.Basin street blues: basin modeling example
7.2.3.Risky business: petroleum prospect risking example
Contents note continued: 7.3.Hands on: reservoir characterization from geophysical data
7.3.1.Black gold in a white plight: reservoir characterization example
7.3.2.Reservoir dogs: seismic and electromagnetic data example
7.4.Hands on: mine planning and safety
7.4.1.I love rock and ore: mining oxide grade example
7.4.2.We will rock you: rock hazard example
7.5.Hands on: groundwater management
7.5.1.Part I: salt water wells in my eyes
groundwater monitoring in Netica
7.5.2.Part II: salt water wells in my eyes
groundwater monitoring in BNT
Appendix: selected statistical models and sampling methods
Appendix A.1 Gaussian distribution
A.1.1.Definition and properties
A.1.2.Decision analysis and VOI results
Appendix A.2 Generalized linear models
A.2.1.Definition and properties
A.2.2.Decision analysis and VOI results
Appendix A.3 Markov chains and hidden Markov models
A.3.1.Definition and properties
Contents note continued: A.3.2.Decision analysis and VOI results
Appendix A.4 Categorical Markov random fields
A.4.1.Definition and properties
A.4.2.Decision analysis and VOI results
Appendix A.5 Discrete graphs and Bayesian networks
A.5.1.Definition and properties
A.5.2.Decision analysis and VOI results
Appendix B Sampling methods.
Presents a unified framework for assessing the value of potential data gathering schemes, with a focus on the Earth sciences.
9781107040267 (hbk.)
2015022572
Earth sciences--Information services.
Geochemical prospecting.
Communication in science.
550.01156 EID / 009925