Geostatistical Analysis of Compositional Data
1st Edition
Published on 3. June 2004
204 pages
978-0-19-803831-3 (ISBN)
Description
Geostatistical Analysis of Compositional Data provides a comprehensive coverage of the theory and practice of analysis of data that have both spatial and compositional dependence, characteristics of most earth science and environmental measurements. No other publication covers this combination of subjects.
More details
Other editions
Additional editions
Vera Pawlowsky-Glahn | Richardo A. Olea
Geostatistical Analysis of Compositional Data
Book
06/2004
Oxford University Press Inc
€149.60
Shipment within 15-20 days
Content
- Intro
- Contents
- 1 Introduction
- 1.1 Statement of the problem
- 1.2 Compositions
- 1.3 Coregionalization
- 1.4 Organization of the book
- 2 Regionalized compositions
- 2.1 First concepts of regionalized compositions
- 2.2 Basis of a regionalized composition
- 2.3 Regionalized subcompositions
- 2.4 Regionalized amalgamations and partitions
- 2.5 alr and clr transformations
- 2.6 Hypothesis of stationarity
- 2.7 The additive logistic normal distribution
- 3 Spatial covariance structure
- 3.1 Second-order stationary case
- 3.1.1 Spurious spatial correlation
- 3.1.2 Defining spatial covariance structure
- 3.1.3 lr autocovariance
- 3.1.4 alr cross-covariance
- 3.1.5 clr cross-covariance
- 3.1.6 Relationships between specifications
- 3.1.7 Symmetry of the spatial covariance structure
- 3.2 Spatial covariance structure under intrinsic hypothesis
- 3.2.1 Intrinsic spatial covariance structure
- 3.2.2 lr semivariogram
- 3.2.3 alr cross-semivariogram
- 3.2.4 clr cross-semivariogram
- 3.2.5 Further relationships between specifications
- 3.3 Spatial covariance structure of an r-basis
- 4 Concepts of null correlation
- 4.1 Null lr cross-correlation
- 4.2 Null Ir autocorrelation
- 4.3 Null alr cross-correlation
- 4.4 Null clr cross-correlation
- 4.5 Composition invariance
- 4.6 Relationship between concepts of null cross-correlation
- 4.7 Subcompositional invariance and partition independence
- 5 Cokriging
- 5.1 The general case of cokriging
- 5.1.1 Cokriging with known mean
- 5.1.2 Cokriging with unknown mean
- 5.2 Normal cokriging
- 5.3 Lognormal cokriging
- 5.3.1 Lognormal cokriging with known mean
- 5.3.2 Lognormal cokriging with unknown mean
- 5.3.3 Comments on lognormal cokriging
- 5.4 alr cokriging
- 5.4.1 alr cokriging with known and unknown mean
- 5.4.2 alr cokriging of a subvector
- 5.4.3 alr autocorrelation and alr cokriging
- 5.4.4 Permutation invariance of alr cokriging estimators
- 5.5 Intrinsic vector random functions
- 6 Practical aspects of compositional data analysis
- 6.1 Dealing with zeros in compositional data
- 6.2 Modeling alr cross-covariance matrices
- 6.3 Exploratory analysis of compositional data
- 6.4 Back transforming alr means and variances
- 6.5 Confidence intervals and confidence regions
- 6.5.1 General concepts
- 6.5.2 Confidence intervals and confidence regions in the nonregionalized case
- 6.5.3 Confidence intervals and confidence regions in the regionalized case
- 6.6 H-data files
- 6.7 Criteria for comparing results
- 6.7.1 Distance between observed samples and estimates
- 6.7.2 STRESS between observed and estimated data set
- 7 Application to real data
- 7.1 The Lyons West oil field of Kansas
- 7.1.1 Description of the field
- 7.1.2 Preparation of the oil-field data
- 7.2 Direct estimation
- 7.3 The alr method
- 7.3.1 The particular case of Lyons West
- 7.3.2 Spatial correlation
- 7.3.3 Comparison of estimation methods
- 7.4 The basis method
- 7.4.1 The basis for Lyons West
- 7.4.2 Modeling of covariances
- 7.4.3 The kriging and cokriging of the basis
- 7.4.4 Explanation of the reversal of the optimal method
- 7.5 Last exercise
- 7.6 Concluding comparisons
- Summary and prospects
- References
- Index
- A
- B
- C
- D
- E
- F
- G
- H
- I
- K
- L
- M
- N
- O
- P
- R
- S
- T
- U
- V
- W
- Z
