這個就不用多說了吧，介紹貝葉斯層次結(jié)構(gòu)模型建模和貝葉斯理論基礎(chǔ)的，費(fèi)了好大勁才找到的，共享一下。 Sudipto Banerjee Bradley P. Carlin Alan E. Gelfand 2004年版 Preface xv 1 Overview of spatial data problems 1 1.1 Introduction to spatial data and models 1 1.1.1 Point-level models 6 1.1.2 Areal models 7 1.1.3 Point process models 8 1.2 Fundamentals of cartography 10 1.2.1 Map projections 10 1.2.2 Calculating distance on the earth's surface 17 1.3 Exercises 18 2 Basics of point-referenced data models 21 2.1 Elements of point-referenced modeling 22 2.1.1 Stationarity 22 2.1.2 Variograms 22 2.1.3 Isotropy 24 2.1.4 Variogram model tting 29 2.2 Spatial process models ? 30 2.2.1 Formal modeling theory for spatial processes 30 2.2.2 Covariance functions and spectra 32 2.2.3 Smoothness of process realizations 36 2.2.4 Directional derivative processes 37 2.2.5 Anisotropy 37 2.3 Exploratory approaches for point-referenced data 39 2.3.1 Basic techniques 39 2.3.2 Assessing anisotropy 44 2.4 Classical spatial prediction 48 2.5 Computer tutorials 52 2.5.1 EDA and variogram tting in S+SpatialStats 52 2.5.2 Kriging in S+SpatialStats 56 2.5.3 EDA, variograms, and kriging in geoR smoothers 75 3.2 Brook's Lemma and Markov random elds 76 3.3 Conditionally autoregressive (CAR) models 79 3.3.1 The Gaussian case 79 3.3.2 The non-Gaussian case 83 3.4 Simultaneous autoregressive (SAR) models 84 3.5 Computer tutorials 88 3.5.1 Adjacency matrix construction in S+SpatialStats 88 3.5.2 SAR and CAR model tting in S+SpatialStats 90 3.5.3 Choropleth mapping using the maps library in S-plus 92 3.6 Exercises 95 4 Basics of Bayesian inference 99 4.1 Introduction to hierarchical modeling and Bayes' Theorem 99 4.2 Bayesian inference 103 4.2.1 Point estimation 103 4.2.2 Interval estimation 104 4.2.3 Hypothesis testing and model choice 105 4.3 Bayesian computation 110 4.3.1 The Gibbs sampler 111 4.3.2 The Metropolis-Hastings algorithm 113 4.3.3 Slice sampling 116 4.3.4 Convergence diagnosis 116 4.3.5 Variance estimation 119 4.4 Computer tutorials 120 4.4