| Non-standard Spatial Statistics | |
| Introduction: Spatial Statistics | p. 3 |
| Individual Versus Ecological Analyses | p. 5 |
| Introduction | p. 5 |
| Spatial Autocorrelation Effects | p. 5 |
| Aggregation Impacts | p. 6 |
| The Syracuse Data | p. 8 |
| Previous Findings for Syracuse | p. 10 |
| Spatial Autocorrelation in the Syracuse Data | p. 11 |
| Spatial Autocorrelation in the Syracuse Data: LN(BLL + 1) Values | p. 11 |
| Spatial Autocorrelation in the Syracuse Data: Appraised House Value | p. 13 |
| Spatial Autocorrelation in the Syracuse Data: Other Sources | p. 16 |
| Bayesian Analysis Using Gibbs Sampling (BUGS) and Model Prediction Experiments | p. 17 |
| Results for the 2000 Census Tracts | p. 20 |
| Discussion and Implications | p. 22 |
| Statistical Models for Spatial Data: Some Linkages and Communalities | p. 25 |
| Introduction | p. 25 |
| Background: Quantifying Spatial Autocorrelation | p. 26 |
| The Moran Scatterplot | p. 27 |
| The Semivariogram Plot | p. 28 |
| Specifications of Spatial Autoregressive and Geostatistical Models | p. 28 |
| Spatial Autoregressive Models | p. 29 |
| Geostatistical Models | p. 31 |
| Linkages Between Spatial Autoregression and Geostatistics | p. 32 |
| A Major Commonality of Spatial Autoregression and Geostatistics | p. 33 |
| Implications for Quantitative Human Geography | p. 35 |
| Frequency Distributions for Simulated Spatially Autocorrelated Random Variables | p. 37 |
| Introduction | p. 37 |
| The Normal Probability Model | p. 38 |
| Simulating Spatially Autocorrelated Normal RVs | p. 39 |
| Simulation Results for an Ideal Regular Hexagonal Surface Partitioning | p. 41 |
| Simulation Results for the China County Geographic Configuration | p. 45 |
| Implications | p. 48 |
| The Poisson Probability Model | p. 50 |
| Simulating Spatially Autocorrelated Poisson RVs | p. 52 |
| Simulation Results for an Ideal Regular Hexagonal Surface Partitioning | p. 55 |
| Simulation Results for the China County Geographic Configuration | p. 56 |
| Implications | p. 60 |
| The Binomial Probability Model, N > 1 | p. 62 |
| Simulating Spatially Autocorrelated Binomial RVs | p. 63 |
| Simulation Results for an Ideal Regular Hexagonal Surface Partitioning | p. 65 |
| Simulation Results for the China County Geographic Configuration | p. 68 |
| Implications | p. 70 |
| Discussion | p. 71 |
| Understanding Correlations Among Spatial Processes | p. 75 |
| Introduction | p. 75 |
| Two Illustrative Examples | p. 75 |
| Geostatistical Semivariogram Model Implications | p. 77 |
| Spatial Autoregressive Model Implications | p. 82 |
| Variance and Covariance Inflation Attributable to Spatial Autocorrelation | p. 85 |
| Effective Sample Size as a Function of ¿X and ¿Y | p. 87 |
| Spatial Filtering Model Implications | p. 89 |
| Correlation Coefficient Decomposition | p. 90 |
| Variance Inflation | p. 93 |
| Discussion | p. 93 |
| Spatially Structured Random Effects: A Comparison of Three Popular Specifications | p. 97 |
| Introduction | p. 97 |
| Modeling Spatial Structure | p. 97 |
| Linear Mixed Models | p. 99 |
| Generalized Linear Mixed Models | p. 105 |
| Degrees of Freedom for GLMM Random Effects | p. 110 |
| Extensions to Space-Time Data Sets | p. 111 |
| Discussion and Implications | p. 114 |
| Spatial Filter Versus Conventional Spatial Model Specifications: Some Comparisons | p. 117 |
| Introduction | p. 117 |
| Background | p. 117 |
| Variation and Covariation Considerations for Poisson Random Variables | p. 120 |
| Heterogeneity in Counts Data | p. 121 |
| Spatial Autocorrelation in Poisson Random Variables | p. 124 |
| Spatial Autocorrelation-induced Correlation Inflation | p. 126 |
| Principal Spatial Statistical Model Specifications | p. 130 |
| The Log-normal Approximation | p. 130 |
| A Winsorized Auto-Poisson Model | p. 131 |
| A Proper CAR Model Specification via GeoBUGS | p. 134 |
| Spatial Filter Model Specifications | p. 136 |
| The Log-normal Approximation Spatial Filter Model | p. 136 |
| A Poisson Spatial Filter Model | p. 137 |
| A Spatial Filter Model Specification via BUGS | p. 139 |
| Discussion | p. 140 |
| Cross-validation Results for the Poisson Spatial Filter Model | p. 141 |
| A Simulation Experiment Based Upon the Poisson Spatial Filter Model | p. 141 |
| Impacts of Incorporating Additional Information | p. 143 |
| Implications for Data Mapping | p. 144 |
| Concluding Comments | p. 147 |
| The Role of Spatial Autocorrelation in Prioritizing Sites Within a Geographic Landscape | p. 151 |
| Introduction: The Problem | p. 151 |
| The Murray Superfund Site: Part I | p. 152 |
| State-of-the-Art Practice | p. 153 |
| A Spatial Methodology: Stage 1, Spatial Sampling Data Collection and Preprocessing | p. 154 |
| The Murray Superfund Site: Part II | p. 156 |
| A Spatial Methodology: Stage 2, Spatial Statistics for Calculating UCLs | p. 159 |
| The Murray Superfund Site: Part III | p. 161 |
| A Spatial Methodology: Stage 3, Prioritizing Subregions for Remediation | p. 163 |
| The Murray Superfund Site: Part IV | p. 163 |
| A Spatial Methodology: Stage 4, Covariation of Contaminants and Joint Pollutant Analyses | p. 164 |
| The Murray Superfund Site: Part V | p. 168 |
| Implications | p. 170 |
| General Conclusions: Spatial Statistics | p. 171 |
| Non-standard Spatial Econometrics | |
| Introduction: Spatial Econometrics | p. 177 |
| A Mixed Linear-Logarithmic Specification for Lotka-Volterra Models with Endogenously Generated SDLS-Variables | p. 179 |
| Lotka-Volterra Models | p. 179 |
| A General Specification | p. 179 |
| Applications | p. 180 |
| Simultaneous Dynamic Least Squares (SDLS) Estimation | p. 181 |
| Mixed Specification | p. 182 |
| Equations | p. 182 |
| Stability | p. 183 |
| Application | p. 183 |
| Conclusion | p. 186 |
| Selecting Spatial Regimes by Threshold Analysis | p. 189 |
| Method | p. 189 |
| Spatial Income Generating Model | p. 190 |
| A Spatial Activity Complex Model | p. 193 |
| Conclusion | p. 197 |
| Appendix | p. 197 |
| Finite Automata | p. 199 |
| A Finite Automaton Bi-regional Dynamic Model | p. 199 |
| An Empirical Application | p. 203 |
| Conclusion | p. 205 |
| Learning from Residuals | p. 207 |
| Residuals | p. 207 |
| Multiple Regimes | p. 209 |
| Spatial Interpolation | p. 212 |
| Composite Parameters | p. 213 |
| Conclusion | p. 215 |
| Verhulst and Poisson Distributions | p. 217 |
| Robust Estimation in the Binary Case: A Linear Logistic Estimator (LLE) | p. 217 |
| A Logistic Dynamic Share Model | p. 219 |
| A Linear Poisson Distribution Estimator | p. 221 |
| Conclusion | p. 225 |
| Qualireg, A Qualitative Regression Method | p. 227 |
| Qualiflex | p. 227 |
| Qualireg | p. 230 |
| Spatial Setting | p. 231 |
| Conclusion | p. 233 |
| Filtering Complexity for Observational Errors and Spatial Bias | p. 235 |
| Complexity, Estimation and Testing | p. 235 |
| Filtering for Observational Errors | p. 237 |
| Further Filtering for Spatial Bias | p. 240 |
| Conclusions | p. 241 |
| General Spatial Econometric Conclusions | p. 243 |
| Epilogue | p. 245 |
| References | p. 247 |
| Author Index | p. 257 |
| Subject Index | p. 259 |
| Table of Contents provided by Ingram. All Rights Reserved. |