I Geostatistical Data
1
Geostatistics
1.1
Targeted Problems
1.2
Spatial Data Example
1.2.1
Main Concepts
1.2.2
Esitimation of Variogram
1.2.3
Decomposition
1.3
Stationary Process
1.3.1
Variogram & Covariogram
1.3.2
Estimation of Variogram & Covariogram
1.3.3
Validity
1.4
Variogram Model Fitting
1.4.1
EDA Check Before Model Fitting
1.4.2
Maximum Likelihood Estimator
1.4.3
MINQ Estimation
2
Spatial Prediction and Kriging
2.1
Scale of variation
2.2
Ordinary Kriging
2.2.1
Model Introduction
2.2.2
Effect of Variogram Estimation
2.3
Cokriging
2.4
Robust Kriging
2.5
Universal Kriging
II Lattice Data
3
Spatial Models on Lattices
3.1
Lattices
3.2
Spatial Models for Lattice Data
3.2.1
Simultaneous Spatial Models
3.2.2
Conditional Spatial Models
3.3
Markov Random Fields
3.3.1
Preparation
3.3.2
Markov Random Field
3.3.3
Hammersley-Clifford Theorem
3.3.4
Pairwise-Only Dependence
3.4
Conditionally Specified Spatial Models
3.4.1
Models for Discrete Data
3.4.2
Models for Continuous Data
3.5
Simultaneously Specified Spatial Models
3.5.1
Spatial Autoregressive Regression Model
3.6
Space-Time Models
3.6.1
STARMA Model
4
Inference for Lattice Models
4.1
Parameter Estimation for Lattice Models
4.1.1
Pseudolikelihood
4.1.2
Gaussian Maximum Likelihood Estimation
4.2
Conditional Gaussian Model
4.3
Auto-logistic Model
4.4
Auto-Poisson Model
III Homework
5
Spatial Analysis Homework
5.1
Overview
5.1.1
Background
5.1.2
Data
5.1.3
Goal
5.2
Exploratory Analysis
5.2.1
Data Preprocessing
5.2.2
Descriptive Statistics
5.3
Model Fitting
5.3.1
Variogram
5.3.2
Kriging Model
5.4
Animation
References
Geostatistics Final Summary
References
Cressie N. Statistics for spatial data[M]. John Wiley & Sons, 2015.
Lecture slides by
Wu Wang
, Renmin University of China