Spatial Analysis Methods and Practice: Describe – Explore – Explain through GIS

Description

This is an introductory textbook on spatial analysis and spatial statistics through GIS. Each chapter presents methods and metrics, explains how to interpret results, and provides worked examples. Topics include: describing and mapping data through exploratory spatial data analysis; analyzing geographic distributions and point patterns; spatial autocorrelation; spatial clustering; geographically weighted regression and OLS regression; and spatial econometrics. The worked examples link theory to practice through a single real-world case study, with software and illustrated guidance. Exercises are solved twice: first through ArcGIS, and then GeoDa. Through a simple methodological framework the book describes the dataset, explores spatial relations and associations, and builds models. Results are critically interpreted, and the advantages and pitfalls of using various spatial analysis methods are discussed. This is a valuable resource for graduate students and researchers analyzing geospatial data through a spatial analysis lens, including those using GIS in the environmental sciences, geography, and social sciences. Contents Preface 1 Think Spatially: Basic Concepts of Spatial Analysis and Space Conceptualization Learning Objectives 1.1 Introduction: Spatial Analysis 1.2 Basic Definitions 1.3 Spatial Data: What Makes Them Special? 1.4 Conceptualization of Spatial Relationships 1.5 Distance Measure 1.5.1 Fixed Distance Band (Sphere of Influence) 1.5.2 Distance Decay 1.6 Contiguity: Adjacency Matrix 1.6.1 Polygons Contiguity 1.6.2 Adjacency Matrix 1.7 Interaction 1.8 Neighborhood and Neighbors 1.8.1 k-Nearest Neighbors (k-NN) 1.8.2 Space–Time Window 1.8.3 Proximity Polygons 1.8.4 Delaunay Triangulation and Triangular Irregular Networks (TIN) 1.9 Spatial Weights and Row Standardization 1.10 Chapter Concluding Remarks Questions and Answers Lab 1 The Project: Spatial Analysis for Real Estate Market Investments Overall Progress Scope of Analysis Dataset Structure Guidelines Section A ArcGIS Exercise 1.1 Getting to Know the Data and Study Region Section B GeoDa Exercise 1.1 Getting to Know the Data and Study Region 2 Exploratory Spatial Data Analysis Tools and Statistics Learning Objectives 2.1 Introduction in Exploratory Spatial Data Analysis, Descriptive Statistics, Inferential Statistics and Spatial Statistics 2.2 Simple ESDA Tools and Descriptive Statistics for Visualizing Spatial Data (Univariate Data) 2.2.1 Choropleth Maps 2.2.2 Frequency Distribution and Histograms 2.2.3 Measures of Center 2.2.4 Measures of Shape 2.2.5 Measures of Spread/Variability – Variation 2.2.6 Percentiles, Quartiles and Quantiles 2.2.7 Outliers 2.2.8 Boxplot 2.2.9 Normal QQ Plot 2.3 ESDA Tools and Descriptive Statistics for Analyzing Two or More Variables (Bivariate Analysis) 2.3.1 Scatter Plot 2.3.2 Scatter Plot Matrix 2.3.3 Covariance and Variance–Covariance Matrix 2.3.4 Correlation Coefficient 2.3.5 Pairwise Correlation 2.3.6 General QQ Plot 2.4 Rescaling Data 2.5 Inferential Statistics and Their Importance in Spatial Statistics 2.5.1 Parametric Methods 2.5.2 Nonparametric Methods 2.5.3 Confidence Interval 2.5.4 Standard Error, Standard Error of the Mean, Standard Error of Proportion and Sampling Distribution 2.5.5 Significance Tests, Hypothesis, p-Value and z-Score 2.6 Normal Distribution Use in Geographical Analysis 2.7 Chapter Concluding Remarks Questions and Answers Lab 2 Exploratory Spatial Data Analysis (ESDA): Analyzing and Mapping Data Overall Progress Scope of the Analysis: Income and Expenses Section A ArcGIS Exercise 2.1 ESDA Tools: Mapping and Analyzing the Distribution of Income Exercise 2.2 Bivariate Analysis: Analyzing Expenditures by Educational Attainment Section B GeoDa Exercise 2.1 ESDA Tools: Mapping and Analyzing the Distribution of Income Exercise 2.2 Bivariate Analysis: Analyzing Expenditures by Educational Attainment 3 Analyzing Geographic Distributions and Point Patterns Learning Objectives 3.1 Analyzing Geographic Distributions: Centrography 3.1.1 Mean Center 3.1.2 Median Center 3.1.3 Central Feature 3.1.4 Standard Distance 3.1.5 Standard Deviational Ellipse 3.1.6 Locational Outliers and Spatial Outliers 3.2 Analyzing Spatial Patterns: Point Pattern Analysis 3.2.1 Definitions: Spatial Process, Complete Spatial Randomness, First- and Second-Order Effects 3.2.2 Spatial Process 3.3 Point Pattern Analysis Methods 3.3.1 Nearest Neighbor Analysis 3.3.2 Ripley’s K Function and the L Function Transformation 3.3.3 Kernel Density Function 3.4 Chapter Concluding Remarks Questions and Answers Lab 3 Spatial Statistics: Measuring Geographic Distributions Overall Progress Scope of the Analysis: Crime Analysis Exercise 3.1 Measuring Geographic Distributions Exercise 3.2 Point Pattern Analysis Exercise 3.3 Kernel Density Estimation Exercise 3.4 Locational Outliers 4 Spatial Autocorrelation Learning Objectives 4.1 Spatial Autocorrelation 4.2 Global Spatial Autocorrelation 4.2.1 Moran’s I Index and Scatter Plot 4.2.2 Geary’s C Index 4.2.3 General G-Statistic 4.3 Incremental Spatial Autocorrelation 4.4 Local Spatial Autocorrelation 4.4.1 Local Moran’s I (Cluster and Outlier Analysis) 4.4.2 Optimized Outlier Analysis 4.4.3 Getis-Ord Gi and Gi* (Hot Spot Analysis) 4.4.4 Optimized Hot Spot Analysis 4.5 Space–Time Correlation Analysis 4.5.1 Bivariate Moran’s I for Space–Time Correlation 4.5.2 Differential Moran’s I 4.5.3 Emerging Hot Spot Analysis 4.6 Multiple Comparisons Problem and Spatial Dependence 4.7 Chapter Concluding Remarks Questions and Answers Lab 4 Spatial Autocorrelation Overall Progress Scope of the Analysis Section A ArcGIS Exercise 4.1 Global Spatial Autocorrelation Exercise 4.2 Incremental Spatial Autocorrelation and Spatial Weights Matrix Exercise 4.3 Cluster and Outlier Analysis (Anselin Local Moran’s I ) Exercise 4.4 Hot Spot Analysis (Getis-Ord Gi*I) and Optimized Hot Spot Analysis Exercise 4.5 Optimized Hot Spot Analysis for Crime Events Section B GeoDa Exercises 4.1 and 4.2 Global Spatial Autocorrelation and Spatial Weights Matrix Exercise 4.3 Cluster and Outlier Analysis (Anselin Local Moran’s I ) Exercise 4.4 Hot Spot Analysis (Getis-Ord Gi*I) 5 Multivariate Data in Geography: Data Reduction and Clustering Learning Objectives 5.1 Multivariate Data Analysis 5.2 Principal Component Analysis (PCA) 5.3 Factor Analysis (FA) 5.4 Multidimensional Scaling (MDS) 5.5 Cluster Analysis 5.5.1 Hierarchical Clustering 5.5.2 k-Means Algorithm (Partitional Clustering) 5.6 Regionalization 5.6.1 SKATER Method 5.6.2 REDCAP Method 5.7 Density-Based Clustering: DBSCAN, HDBSCAN, OPTICS 5.8 Similarity Analysis: Cosine Similarity 5.9 Chapter Concluding Remarks Questions and Answers Lab 5 Multivariate Statistics: Clustering Overall Progress Scope of the Analysis Section A ArcGIS Exercise 5.1 k-Means Clustering Exercise 5.2 Spatial Clustering (Regionalization) Exercise 5.3 Similarity Analysis Exercise 5.4 Synthesis Section B GeoDa Exercise 5.1 k-Means Clustering Exercise 5.2 Spatial Clustering 6 Modeling Relationships: Regression and Geographically Weighted Regression Learning Objectives 6.1 Simple Linear Regression 6.1.1 Simple Linear Regression Assumptions 6.1.2 Ordinary Least Squares (Intercept and Slope by OLS) 6.2 Multiple Linear Regression (MLR) 6.2.1 Multiple Regression Basics 6.2.2 Model Overfit: Selecting the Number of Variables by Defining a Functional Relationship 6.2.3 Missing Values 6.2.4 Outliers and Leverage Points 6.2.5 Dummy Variables 6.2.6 Methods for Entering Variables in MLR: Explanatory Analysis; Identifying Causes and Effects 6.3 Evaluating Linear Regression Results: Metrics, Tests and Plots 6.3.1 Multiple r 6.3.2 Variation and Coefficient of Determination R-Squared 6.3.3 Adjusted R-Squared 6.3.4 Predicted R-Squared 6.3.5 Standard Error (Deviation) of Regression (or Standard Error of the Estimate) 6.3.6 F-Test of the Overall Significance 6.3.7 t-Statistic (Coefficients’ Test) 6.3.8 Wald Test (Coefficient’s Test) 6.3.9 Standardized Coefficients (Beta) 6.3.10 Residuals, Residual Plots and Standardized Residuals 6.3.11 Influential Points: Outliers and High-Leverage Observations 6.4 Multiple Linear Regression Assumptions: Diagnose and Fix 6.5 Multicollinearity 6.6 Worked Example: Simple and Multiple Linear Regression 6.7 Exploratory Regression 6.8 Geographically Weighted Regression 6.8.1 Spatial Kernel Types 6.8.2 Bandwidth 6.8.3 Interpreting GWR Results and Practical Guidelines 6.9 Chapter Concluding Remarks Questions and Answers Lab 6 OLS, Explanatory Regression, GWR Overall Progress Scope of the Analysis Exercise 6.1 Exploratory Regression Exercise 6.2 OLS Regression Exercise 6.3 GWR 7 Spatial Econometrics Learning Objectives 7.1 Spatial Econometrics 7.2 Spatial Dependence: Spatial Regression Models and Diagnostics 7.2.1 Diagnostics for Spatial Dependence 7.2.2 Selecting between Spatial Lag or Spatial Error Model 7.2.3 Estimation Methods 7.3 Spatial Lag Model 7.3.1 Spatial Two-Stage Least Squares (S2SLS) 7.3.2 Maximum Likelihood 7.4 Spatial Error Model 7.5 Spatial Filtering 7.6 Spatial Heterogeneity: Spatial Regression Models 7.7 Spatial Regimes 7.8 Chapter Concluding Remarks Questions and Answers Lab 7 Spatial Econometrics Overall Progress Scope of the Analysis Exercise 7.1 OLS Exercise 7.2 Spatial Error Model Exercise 7.3 OLS with Spatial Regimes Exercise 7.4 Spatial Error by Spatial Regimes References Index