ebook img

Exploratory and explanatory statistical analysis of spatial data PDF

270 Pages·1979·7.032 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Exploratory and explanatory statistical analysis of spatial data

Exploratory and explanatory statistical analysis of spatial data Exploratory and explanatory statistical analysis of spatial data CORNELIS P. A. BARTELS and RONALD H. KETELLAPPER, editors University of Groningen GMartinus GJVijhoff Publishing Boston/TheHague/London 1979 Distributors for North America: Martinus Nijhoff Publishing Kluwer Boston, Inc. 160 Old Derby Street Hingham, Massachusetts 02043 Distributors outside North America: K1uwer Academic Publishers Group Distribution Centre P.O. Box 322 3300 AH Dordrecht, The Netherlands Library of Congress Cataloging in Publication Data Regional Science Symposium, University of Groningen, 1977. Exploratory and explanatory statistical analysis of spatial data. Contains, for the most part, papers presented at a Regional Science Symposium, held at the Faculty of Economics of the University of Groningen in the Netherlands. Includes bibliographies. I. Regional economics-Statistical methods-Con gresses. 2. Regional planning-Statistical methods Congresses. I. Bartels, Comelis, P.A. II. Ketellapper, Ronald H. III. Title. HT391.R3374 1977 330'.01'82 79-13142 ISBN-13: 978-94-009-9235-1 e-ISBN-13: 978-94-009-9233-7 001: 10.10071978-94-009-9233-7 Copyright ©1979 by Martinus Nijhoff Publishing. Softcover reprint of the hardcover 15t edition 1979 No part of this book may be reproduced in any form by print, photoprint, microfilm or any other means, without written permission from the publisher. Preface In September 1977 a "Regional Science Symposium" was held at the Faculty of Economics of the University of Goningen in the Netherlands. The impetus in organizing this symposium was the recent estab lishmen t at the F acuIty of Economics of a group engaged in teaching and research within the field of regional science. The aim of the symposium was to familiarize university members with regional science and to introduce the new group to both the national and international scene. Two separate topics of potential interest to both researchers and policy-makers were selected. The first theme, spatial inequalities and regional development, was chosen because of its central place in regional science. Authors from several disciplines were asked to approach this theme from a general, policy-oriented point of view. This ensured the spotlighting of the various dimensions of spatial inequality and its implications for regional policy. The results of their efforts have been collected in a volume entitled Spatial Inequalities and Regional Development. The second theme focussed on spatial statistical analysis. This branch of statistics is a relatively new one. It is receiving growing attention from researchers in the field of applied regional science. The conference dealing with this topic concentrated on recent research results related to the use of appropriate statistical and econometric methods for analyzing spatial data. The papers con cerned have been collected in another volume, entitled Exploratory and Explanatory Statistical Analysis of Spatial Data. Both volumes contain, for the most part, papers presented at the symposium. Some additional papers have been included to improve the consistency of the volumes. All contributions have been revised prior to final publication. In this process critical comments made by participants at the symposium have proven to be very helpful. We believe that these efforts have helped considerably to improve the quality of both volumes. Groningen, April 1978 The editors Contents PREFACE v CONTENTS VII LIST OF CONTRIBUTORS XI Part 1: Introduction 1 1. GENERAL INTRODUCTION 3 CORNELIS P.A. BARTELS AND RONALD H. KETELLAPPER 2. OPERATIONAL STATISTICAL METHODS FOR ANALYSING SPATIAL DATA 5 CORNELIS P. A. BARTELS 2.1. Introduction 5 2.2. The structure of spatial data 6 2.3. Methods based on simple correlations between cross-regional data 7 2.4. Time-series analysis applied to spatial data 13 2.5. Adaptations of time-series analysis to the spatial context 20 2.6. Single equation explanatory models 28 2.7. Simultaneous equation models with spatial data 40 2.8. Some remaining topics 42 2.9. Final remarks 45 References 45 VIII CONTENTS Part 2: Exploratory statistical analysis 51 3. THE ANALYSIS OF GEOGRAPHICAL MAPS 53 BRIAN D. RIPLEY 3.1. Introduction 53 3.2. Methods of analysis 54 3.3. Models 55 3.4. Tests for randomness 60 3.5. Examples 65 3.6. Conclusions 67 References 71 4. CONSTRUCTION OF INTERREGIONAL INPUT- 73 OUTPUT TABLES BY EFFICIENT INFORMATION ADDING FOLKE SNICKARS 4.1. Introduction 73 4.2. Regional and national accounts 75 4.3. Generation of survey-based transaction tables 79 4.4. Results of the statistical estimations 91 4.5. Results of the minimum information estimations 105 4.6. Some conclusions 110 References III 5. FURTHER EVIDENCE ON ALTERNATIVE PRO CEDURES FOR TESTING OF SPATIAL AUTO- CORRELATION AMONG REGRESSION DIS- 113 TURBANCES ANDRIES S. BRANDSMA AND RONALD H. KETELLAPPER 5.1. Introduction 113 5.2. Formulation of the statistical decision problem 115 5.3. Moran's test statistic 116 5.4. Moments of the Moran statistic using OLS and LUS estimators 118 5.5. The likelihood ratio test 122 5.6. Simulation study of the Moran and likelihood ratio tests 125 5.7. Results 126 CONTENTS IX 5.8. Conclusions 132 Appendix 134 References 135 Part 3: Explanatory statistical analysis 137 6. MULTIVARIATE MODELS OF DEPENDENT SPATIAL DATA 139 BERND STREITBERG 6.1. Introduction 139 6.2. Decomposable covariance structures 141 6.3. Linear models 157 6.4. Principal components 172 6.5. Conclusion 173 References 176 7. BAYESIAN ANALYSIS OF THE LINEAR MODEL WITH SPATIAL DEPENDENCE 179 LESLIE W. HEPPLE 7.1. Introduction 179 7.2. The nature of Bayesian inference 180 7.3. Linear regression model with spatially auto- correlated disturbances 185 7.4. An empirical application 193 7.5. Concluding remarks 197 References 198 8. ALTERNATIVE METHODS OF ESTIMATING SPATIAL INTERACTION MODELS AND THEIR PERFORMANCE IN SHORT-TERM FORECASTING 201 STAN OPENSHAW 8.1. Introduction 201 8.2. Description of data and models 203 8.3. Parameter estimation and model calibration in terms of 1966 and 1971 data 206 8.4. On the accuracy of short-term forecasts made by spatial interaction models 212 X CONTENTS 8.5. An evaluation of some alternatives designed to improve model performance 216 8.6. Conclusions 223 References 224 9. TWO ESTIMATION METHODS FOR SINGLY CON- STRAINED SPATIAL DISTRIBUTION MODELS 227 JAN VAN EST AND JAN VAN SETTEN 9.1. Introduction 227 9.2. The calibration of a model 228 9.3. The maximum likelihood method 230 9.4. The least-squares method for the singly con- strained model 234 9.5. Numerical results 237 9.6. Conclusions 240 References 241 Part 4: The introduction of stochastics in regional control 243 10. STOCHASTIC CONTROL OF REGIONAL ECONOMIES 245 ROBERT J. BENNETT AND K.C. TAN 10.1. Introduction 245 10.2. Mathematical representation of regional systems 246 10.3. Optimal control models of regional systems 250 10.4 Interaction of optimal control of regional economies with national governments 255 10.5. Problems in applying optimal control to regional systems 264 10.6. Conclusion 266 References 267 List of contributors Cornelis P.A. Bartels is assistant professor in regional economics at the University of Groningen (Netherlands). His publications include a book on economic aspects of regional welfare and several articles on development economics, income distribu tion, regional unemployment and econometric techniques applied to spatial data. Robert J. Bennett is affiliated with the Department of Geography of the University of Cambridge. His various publications include books on spatial time series and en vironmental systems, and articles on the identification, representation and estimation of dynamic spatial models, optimal control models of regional economies, and tech niques for non-stationary parameter estimation. Andries S. Brandsma was student in the Faculty of Econometrics at the University of Groningen, Netherlands. His master thesis focused on methods to account for spatial autocorrelation in the estimation of regional economic models. Leslie W. Hepple is at the Department of Geography of the University of Bristol, England. He published studies on several aspects of econometric estimation with spatial data, e.g., the use of stochastic process theory in spatial analYSiS, spectral analysis, maximum likelihood estimation and Bayesian analysis of regional models. Ronald H. Ketellapper is assistant professor in econometrics at the University of Groningen (Netherlands). His research is mainly in the field of the estimation of eco nometric models, the analysis of errors-in-variables models, and the study of tech niques to test for spatial autocorrelation. Stan Openshaw is lecturer at the Department of Town and Country Planning of the University of Newcastle upon Tyne, England. His research concentrated on several aspects of the use of spatial interaction models in planning, especially the specification of deterrence functions, the determination of optimal zonings, and properties of alternative specifications of such interaction models. Brian D. Ripley is associated with the Department of Mathematics of the Imperial College in London. He has published articles on the theoretical and empirical analysis of stationary point process in space. Folke Snickars is member of the research group for Urban and Regional Planning at XII LIST OF CONTRIBUTORS the Royal Institute of Technology in Stockholm. His research includes studies on the application of information theory in regional science, and the analysis of regional migration. Bernd Streitberg works in the field of econometrics and statistics at the Freie Univer sitat, Berlin. His research concentrated on the estimation oflinear multivariate models with dependent data. K.C. Tan worked at the Department of Geography of the University College, London. He published on optimal control theory for linear econometric systems with linear equality and inequality constraints on the control variables. Jan van Est is research fellow at TNO, Delft, Netherlands. He works on the estimation and application of large scale spatial distribution models. Jan van Setten is research fellow at TNO, Delft, Netherlands. His research includes the estimation of spatial interaction models; and the specification of mathematical models of spatial patterns in the service system.

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.