文档介绍：An Introduction to Spatial Regression
Models in the Social Sciences
Michael D. Ward
Department of Political Science
University of Washington
Seattle, Washington, USA
******@
and
Kristian Skrede Gleditsch
Department of Government
University of Essex
Colchester, UK
******@
c 2007 Michael D. Ward & Kristian Skrede Gleditsch
June 15, 2007, 4:10 am PST
To the accident that is geography . . .
Contents
Preface v
Chapter 1. Introduction 1
1. Interaction and Social Science 1
2. Democracy Around the World 3
3. Introducing Spatial Dependence 7
4. Maps as Visual Displays of Data 9
5. Measuring Spatial Association and Correlation 11
6. Measuring Proximity 16
7. Estimating Spatial Models 23
8. Summary 27
Chapter 2. Spatially Lagged Dependent Variables 29
1. Regression with Spatially Lagged Dependent Variables 29
2. Estimating the Spatially Lagged y Model 34
3. MLE Estimates of the Spatially Lagged Y Model of Democracy 36
4. Equilibrium Effects in the Spatially Lagged y Model 37
5. Spatial Dependence in Turnout in Italy 41
6. Using Different Weights Matrices in a SLDV model 46
7. The Spatially Lagged Dependent Variable vs. OLS with Dummy Variables 50
Chapter 3. Spatial Error Model 55
1. Introduction 55
2. The Spatial Error Model 55
3. MLE Estimation of the Spatial Errors Model 57
4. Example: Democracy and Development 57
5. Spatially Lagged y vs. Spatial Errors 58
6. Assessing Spatial Error in Dyadic Trade Flows 59
7. Conclusion 64
Chapter 4. Extensions 65
1. Introduction 65
2. Specifying Connectivities 65
3. Inference and Model Evaluation 68
4. Summary 72
Appendix A. Software Options 73
iii
iv CONTENTS
Appendix. Bibliography 75
PREFACE v
Preface
Spatial ideas can make substantial contributions to social science research. This
book provides a self-contained introduction for social scientists of how the analysis of
spatial dependence can be integrated into a regres