This thesis explores whether scanner data can be used to inform Consumer Price Index (CPI) construction, with particular reference to the issues of substitution bias and choice of aggregation dimensions. The potential costs and benefits of using scanner data are reviewed. Existing estimates of substitution bias are found to show considerable variation. An Australian scanner data set is used to estimate substitution bias for six different aggregation methods and for fixed base and superlative indexes. Direct and chained indexes are also calculated. Estimates of substitution bias are found to be highly sensitive to both the method of aggregation used and whether direct or chained indexes were used. The ILO (2004) recommends the use of dissimilarity indexes to determine the issue of when to chain. This thesis provides the first empirical study of dissimilarity indexes in this context. The results indicate that dissimilarity indexes may not be sufficient to resolve the issue. A Constant Elasticity of Substitution (CES) index provides an approximate estimate of substitution-bias-free price change, without the need for current period expenditure weights. However, an elasticity parameter is needed. Two methods, referred to as the algebraic and econometric methods, were used to estimate the elasticity parameter. The econometric approach involved the estimation of a system of equations proposed by Diewert (2002a). This system has not been estimated previously. The results show a relatively high level of substitution at the elementary aggregate level, which supports the use a Jevons index, rather than Carli or Dutot indexes, at this level. Elasticity parameter estimates were found to vary considerably across time, and statistical testing showed that elasticity parameter estimates were significantly different across estimation methods. Aggregation is an extremely important issue in the compilation of the CPI. However, little information exists about 'appropriate' aggregation methods. Aggregation is typically recommended over 'homogenous' units. An hedonic framework is used to test for item homogeneity across four supermarket chains and across all stores within each chain. This is a novel approach. The results show that treating the same good as homogenous across stores which belong to the same chain may be recommended.

This thesis explores whether scanner data can be used to inform Consumer Price Index (CPI) construction, with particular reference to the issues of substitution bias and choice of aggregation dimensions. The potential costs and benefits of using scanner data are reviewed. Existing estimates of substitution bias are found to show considerable variation. An Australian scanner data set is used to estimate substitution bias for six different aggregation methods and for fixed base and superlative indexes. Direct and chained indexes are also calculated. Estimates of substitution bias are found to be highly sensitive to both the method of aggregation used and whether direct or chained indexes were used. The ILO (2004) recommends the use of dissimilarity indexes to determine the issue of when to chain. This thesis provides the first empirical study of dissimilarity indexes in this context. The results indicate that dissimilarity indexes may not be sufficient to resolve the issue. A Constant Elasticity of Substitution (CES) index provides an approximate estimate of substitution-bias-free price change, without the need for current period expenditure weights. However, an elasticity parameter is needed. Two methods, referred to as the algebraic and econometric methods, were used to estimate the elasticity parameter. The econometric approach involved the estimation of a system of equations proposed by Diewert (2002a). This system has not been estimated previously. The results show a relatively high level of substitution at the elementary aggregate level, which supports the use a Jevons index, rather than Carli or Dutot indexes, at this level. Elasticity parameter estimates were found to vary considerably across time, and statistical testing showed that elasticity parameter estimates were significantly different across estimation methods. Aggregation is an extremely important issue in the compilation of the CPI. However, little information exists about 'appropriate' aggregation methods. Aggregation is typically recommended over 'homogenous' units. An hedonic framework is used to test for item homogeneity across four supermarket chains and across all stores within each chain. This is a novel approach. The results show that treating the same good as homogenous across stores which belong to the same chain may be recommended.

by Vera Hau Tsz Lai. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1996. / Includes bibliographical references (leaves 227-236). / ABSTRACT --- p.i-ii / ACKNOWLEDGEMENTS --- p.iii-iv / TABLE OF CONTENTS --- p.v-ix / LIST OF FIGURES --- p.x / LIST OF PLATES --- p.xi-xiii / LIST OF TABLES --- p.xiv-xvi / Chapter CHAPTER I --- INTRODUCTION --- p.1 / Chapter 1.1 --- Problem Statement --- p.1 / Chapter 1.2 --- Role of GIS in Housing Price Study --- p.3 / Chapter 1.3 --- Research Objectives --- p.4 / Chapter 1.4 --- Significance --- p.5 / Chapter 1.5 --- Methodologies --- p.6 / Chapter 1.6 --- Organization of Thesis --- p.7 / Chapter CHAPTER II --- LITERATURE REVIEW --- p.9 / Chapter 2.1 --- Introduction --- p.9 / Chapter 2.2 --- Geography of Housing --- p.10 / Chapter 2.3 --- Housing as a Research Question --- p.11 / Chapter 2.4 --- Housing Services and Housing Price --- p.12 / Chapter 2.5 --- Property Price Valuation --- p.14 / Chapter 2.6 --- Hedonic Price Function --- p.15 / Chapter 2.6.1 --- Dependent Variable - Property Price --- p.16 / Chapter 2.6.2 --- Independent Variables Affecting Housing Price --- p.17 / Chapter 2.6.2.1 --- Aspatial Factors --- p.17 / Chapter 2.6.2.2 --- Spatial Factors --- p.18 / Chapter 2.6.2.3 --- Evaluation on Importance of Parameters --- p.26 / Chapter 2.7 --- Functional Form of Hedonic Price Models --- p.33 / Chapter 2.7.1 --- Conventional Specifications --- p.34 / Chapter 2.7.2 --- Box-Cox Transformation --- p.34 / Chapter 2.7.3 --- Conventional Specifications versus Box-Cox Transformation --- p.35 / Chapter 2.8 --- Submarket Analysis and its Delineation --- p.36 / Chapter 2.9 --- Geographic Information Systems --- p.39 / Chapter 2.10 --- GIS in Real Estate --- p.39 / Chapter 2.11 --- Present Adoption of GIS in Real Estate --- p.42 / Chapter 2.11.1 --- Commercial Applications --- p.42 / Chapter 2.11.2 --- Research-wise Applications --- p.43 / Chapter 2.12 --- Hedonic Price Study with GIS --- p.43 / Chapter 2.13 --- Conclusion --- p.45 / Chapter CHAPTER III --- THE STUDY AREA AND RESEARCH METHODOLOGY --- p.47 / Chapter 3.1 --- Introduction --- p.47 / Chapter 3.2 --- Real Estate Sector in Hong Kong --- p.47 / Chapter 3.2.1 --- Importance to Local Economy --- p.48 / Chapter 3.2.2 --- Importance to Housing Production --- p.48 / Chapter 3.3 --- Urban Development and Housing in Hong Kong --- p.51 / Chapter 3.3.1 --- Land Availability and Landuses --- p.51 / Chapter 3.3.2 --- Housing and Urban Development --- p.54 / Chapter 3.3.2.1 --- Early Period of Industrialization --- p.54 / Chapter 3.3.2.2 --- Phase of Economic Restructuring --- p.55 / Chapter 3.3.3 --- Urban Renewal --- p.55 / Chapter 3.3.4 --- Comprehensive Housing Projects --- p.56 / Chapter 3.4 --- New Town Housing - Public or Private-Led --- p.57 / Chapter 3.5 --- Hedonic Price of Private Dormitory in Hong Kong --- p.61 / Chapter 3.5.1 --- Temporal Change in Property Price --- p.62 / Chapter 3.5.2 --- Spatial Variation of Property Price --- p.66 / Chapter 3.6 --- The Research --- p.68 / Chapter 3.6.1 --- Cartographic Analysis --- p.68 / Chapter 3.6.2 --- Hedonic Price Model --- p.69 / Chapter 3.6.3 --- Dependent Variable --- p.69 / Chapter 3.6.4 --- Independent Variables --- p.70 / Chapter 3.6.5 --- Chosen Functional Form in this Research --- p.72 / Chapter 3.6.6 --- Submarket Analysis in Hong Kong --- p.72 / Chapter 3.7 --- Conclusion --- p.72 / Chapter CHAPTER IV --- DATABASE CONSTRUCTIONS --- p.74 / Chapter 4.1 --- Introduction --- p.74 / Chapter 4.2 --- Data Collection --- p.74 / Chapter 4.2.1 --- Base Maps --- p.75 / Chapter 4.2.2 --- Housing Stock and its Attributes --- p.76 / Chapter 4.2.3 --- Official Statistics --- p.76 / Chapter 4.2.4 --- School Quality --- p.77 / Chapter 4.3 --- Data Input --- p.78 / Chapter 4.3.1 --- Graphical Input --- p.78 / Chapter 4.3.1.1 --- Base Maps --- p.78 / Chapter 4.3.1.2 --- Line Data --- p.78 / Chapter 4.3.1.3 --- Point/Polygon Data --- p.79 / Chapter 4.3.2 --- Attribute Data Input --- p.82 / Chapter 4.4 --- Data Editing and Conversions --- p.82 / Chapter 4.4.1 --- Graphical Input --- p.82 / Chapter 4.4.1.1 --- Standard Coverage Editing Procedures --- p.82 / Chapter 4.4.1.2 --- Specific Coverage Editing Procedures --- p.83 / Chapter 4.4.2 --- Attribute Data --- p.84 / Chapter 4.4.2.1 --- Housing Attributes --- p.84 / Chapter 4.4.2.2 --- Landuse Mix --- p.88 / Chapter 4.4.2.3 --- Socioeconomic Status --- p.91 / Chapter 4.4.2.4 --- Employment Figures --- p.91 / Chapter 4.5 --- Data Pre-processing and Manipulation --- p.93 / Chapter 4.5.1 --- Employment Potentials --- p.93 / Chapter 4.5.2 --- Socioeconomic Variables --- p.96 / Chapter 4.5.2.1 --- Interpretation --- p.97 / Chapter 4.5.3 --- School Quality --- p.107 / Chapter 4.5.4 --- Proximity Measurements --- p.110 / Chapter 4.5.5 --- Final Step of Association : Overlay Operations --- p.110 / Chapter 4.6 --- Conclusion --- p.112 / Chapter CHAPTER V --- CARTOGRAPHIC ANALYSIS --- p.114 / Chapter 5.1 --- Introduction --- p.114 / Chapter 5.2 --- Representation of Data --- p.114 / Chapter 5.2.1 --- Location of Premises --- p.114 / Chapter 5.2.2 --- Proximity --- p.118 / Chapter 5.2.3 --- School Quality --- p.118 / Chapter 5.2.4 --- Landuse Mix --- p.129 / Chapter 5.2.5 --- Employment --- p.132 / Chapter 5.2.6 --- Property Price --- p.137 / Chapter 5.3 --- Results and Discussions --- p.137 / Chapter 5.3.1 --- Temporal Variation on Housing Supply --- p.143 / Chapter 5.3.2 --- Temporal Variation on Floor Size --- p.145 / Chapter 5.3.3 --- Temporal Variation on Property Price --- p.148 / Chapter 5.4 --- Locational Variations --- p.150 / Chapter 5.4.1 --- Shift towards the New Towns --- p.150 / Chapter 5.4.2 --- Relative Importance among Districts in New Towns --- p.154 / Chapter 5.4.3 --- Pattern of Development --- p.158 / Chapter 5.4.3.1 --- Urban Core --- p.158 / Chapter 5.4.3.2 --- New Towns --- p.161 / Chapter 5.5 --- Spatial Variations on Floor Size --- p.171 / Chapter 5.6 --- Spatial Variations on Property Price --- p.176 / Chapter 5.7 --- Conclusion --- p.181 / Chapter CHAPTER VI --- STATISTICAL ANALYSIS --- p.183 / Chapter 6.1 --- Introduction --- p.183 / Chapter 6.2 --- The Data Set --- p.183 / Chapter 6.3 --- Stepwise Regression Modeling --- p.184 / Chapter 6.4 --- Correlation among Variables --- p.184 / Chapter 6.5 --- Validation of the Models --- p.186 / Chapter 6.6 --- Findings --- p.193 / Chapter 6.6.1 --- Pooled Market Results --- p.193 / Chapter 6.6.2 --- Submarket Level Analyses --- p.198 / Chapter 6.6.2.1 --- "Small-Sized, Low-Priced Flats " --- p.200 / Chapter 6.6.2.2 --- "Small-Sized, High-Priced Flats " --- p.203 / Chapter 6.6.2.3 --- "Medium-Sized, Low-Priced Flats " --- p.206 / Chapter 6.6.2.4 --- "Medium-Sized, High-Priced Flats " --- p.210 / Chapter 6.6.2.5 --- "Large-Sized, High-Priced Flats " --- p.213 / Chapter 6.7 --- Conclusion --- p.213 / Chapter CHAPTER VII --- CONCLUSION --- p.217 / Chapter 7.1 --- Summary of Findings --- p.217 / Chapter 7.1.1 --- Summary on Housing Development in Hong Kong…… --- p.217 / Chapter 7.1.2 --- Summary from Hedonic Price Models --- p.220 / Chapter 7.1.3 --- Significance of GIS --- p.222 / Chapter 7.2 --- Limitations and Recommendations --- p.222 / Chapter 7.3 --- Direction of Future Research --- p.226 / BIBLIOGRAPHY --- p.227 / APPENDICES --- p.237 / APPENDIX 1 --- p.238 / District Map of Hong Kong --- p.239 / APPENDIX II --- p.240 / List of Districts and its Components --- p.241 / APPENDIX III --- p.243 / Tertiary Planning Units (TPUs) - District Conversion List --- p.244