A comprehensive guide to implementing SAE methods for poverty studies and poverty mapping There is an increasingly urgent demand for poverty and living conditions data, in relation to local areas and/or subpopulations.
A comprehensive guide to implementing SAE methods for poverty studies and poverty mapping There is an increasingly urgent demand for poverty and living conditions data, in relation to local areas and/or subpopulations. Policy makers and stakeholders need indicators and maps of poverty and living conditions in order to formulate and implement policies, (re)distribute resources, and measure the effect of local policy actions. Small Area Estimation (SAE) plays a crucial role in producing statistically sound estimates for poverty mapping. This book offers a comprehensive source of information regarding the use of SAE methods adapted to these distinctive features of poverty data derived from surveys and administrative archives. The book covers the definition of poverty indicators, data collection and integration methods, the impact of sampling design, weighting and variance estimation, the issue of SAE modelling and robustness, the spatio-temporal modelling of poverty, and the SAE of the distribution function of income and inequalities. Examples of data analyses and applications are provided, and the book is supported by a website describing scripts written in SAS or R software, which accompany the majority of the presented methods. Key features:
A comprehensive guide to implementing SAE methods for poverty studies and poverty mapping There is an increasingly urgent demand for poverty and living conditions data, in relation to local areas and/or subpopulations. Policy makers and stakeholders need indicators and maps of poverty and living conditions in order to formulate and implement policies, (re)distribute resources, and measure the effect of local policy actions. Small Area Estimation (SAE) plays a crucial role in producing statistically sound estimates for poverty mapping. This book offers a comprehensive source of information regarding the use of SAE methods adapted to these distinctive features of poverty data derived from surveys and administrative archives. The book covers the definition of poverty indicators, data collection and integration methods, the impact of sampling design, weighting and variance estimation, the issue of SAE modelling and robustness, the spatio-temporal modelling of poverty, and the SAE of the distribution function of income and inequalities. Examples of data analyses and applications are provided, and the book is supported by a website describing scripts written in SAS or R software, which accompany the majority of the presented methods. Key features : Presents a comprehensive review of SAE methods for poverty mapping Demonstrates the applications of SAE methods using real-life case studies Offers guidance on the use of routines and choice of websites from which to download them Analysis of Poverty Data by Small Area Estimation offers an introduction to advanced techniques from both a practical and a methodological perspective, and will prove an invaluable resource for researchers actively engaged in organizing, managing and conducting studies on poverty.
A comprehensive guide to implementing SAE methods for poverty studies and poverty mapping There is an increasingly urgent demand for poverty and living conditions data, in relation to local areas and/or subpopulations. Policy makers and stakeholders need indicators and maps of poverty and living conditions in order to formulate and implement policies, (re)distribute resources, and measure the effect of local policy actions. Small Area Estimation (SAE) plays a crucial role in producing statistically sound estimates for poverty mapping. This book offers a comprehensive source of information regarding the use of SAE methods adapted to these distinctive features of poverty data derived from surveys and administrative archives. The book covers the definition of poverty indicators, data collection and integration methods, the impact of sampling design, weighting and variance estimation, the issue of SAE modelling and robustness, the spatio-temporal modelling of poverty, and the SAE of the distribution function of income and inequalities. Examples of data analyses and applications are provided, and the book is supported by a website describing scripts written in SAS or R software, which accompany the majority of the presented methods. Key features : Presents a comprehensive review of SAE methods for poverty mapping Demonstrates the applications of SAE methods using real-life case studies Offers guidance on the use of routines and choice of websites from which to download them Analysis of Poverty Data by Small Area Estimation offers an introduction to advanced techniques from both a practical and a methodological perspective, and will prove an invaluable resource for researchers actively engaged in organizing, managing and conducting studies on poverty.
Monica Pratesi, Department of Economics and Management, University of Pisa, Italy.
Monica's research field includes small area estimation, inference in elusive populations, nonresponse, design effect in fitting statistical models. Monica is currently involved as researcher and reference person of the DEM-UNIPI in the project EFRAME(European FRAmework for MEasuring progress) funded under the 7th FP (eframeproject.eu/).
Foreword xv Preface xvii Acknowledgements xxiii About the Editor xxv List of Contributors xxvii 1 Introduction on Measuring Poverty at Local Level Using Small Area Estimation Methods 1
Monica Pratesi and Nicola Salvati 1.1 Introduction 1 1.2 Target Parameters 2 1.2.1 Definition of the Main Poverty Indicators 2 1.2.2 Direct and Indirect Estimate of Poverty Indicators at Small Area Level 3 1.3 Data-related and Estimation-related Problems for the Estimation of Poverty Indicators 5 1.4 Model-assisted and Model-based Methods Used for the Estimation of Poverty Indicators: a Short Review 7 1.4.1 Model-assisted Methods 7 1.4.2 Model-based Methods 12 References 15 Part I DEFINITION OF INDICATORS AND DATA COLLECTION AND INTEGRATION METHODS 2 Regional and Local Poverty Measures 21
Achille Lemmi and Tomasz Panek 2.1 Introduction 21 2.2 Poverty – Dilemmas of Definition 22 2.3 Appropriate Indicators of Poverty and Social Exclusion at Regional and Local Levels 23 2.3.1 Adaptation to the Regional Level 23 2.4 Multidimensional Measures of Poverty 25 2.4.1 Multidimensional Fuzzy Approach to Poverty Measurement 25 2.4.2 Fuzzy Monetary Depth Indicators 26 2.5 Co-incidence of Risks of Monetary Poverty and Material Deprivation 30 2.6 Comparative Analysis of Poverty in EU Regions in 2010 31 2.6.1 Data Source 31 2.6.2 Object of Interest 31 2.6.3 Scope and Assumptions of the Empirical Analysis 32 2.6.4 Risk of Monetary Poverty 32 2.6.5 Risk of Material Deprivation 33 2.6.6 Risk of Manifest Poverty 37 2.7 Conclusions 38 References 39 3 Administrative and Survey Data Collection and Integration 41
Alessandra Coli, Paolo Consolini and Marcello D'Orazio 3.1 Introduction 41 3.2 Methods to Integrate Data from Different Data Sources: Objectives and Main Issues 43 3.2.1 Record Linkage 43 3.2.2 Statistical Matching 46 3.3 Administrative and Survey Data Integration: Some Examples of Application in Well-being and Poverty Studies 50 3.3.1 Data Integration for Measuring Disparities in Economic Well-being at the Macro Level 51 3.3.2 Collection and Integration of Data at the Local Level 53 3.4 Concluding Remarks 56 References 57 4 Small Area Methods and Administrative Data Integration 61
Li-Chun Zhang and Caterina Giusti 4.1 Introduction 61 4.2 Register-based Small Area Estimation 63 4.2.1 Sampling Error: A Study of Local Area Life Expectancy 63 4.2.2 Measurement Error due to Progressive Administrative Data 65 4.3 Administrative and Survey Data Integration 68 4.3.1 Coverage Error and Finite-population Bias 68 4.3.2 Relevance Error and Benchmarked Synthetic Small Area Estimation 70 4.3.3 Probability Linkage Error 75 4.4 Concluding Remarks 80 References 81 Part II IMPACT OF SAMPLING DESIGN, WEIGHTING AND VARIANCE ESTIMATION 5 Impact of Sampling Designs in Small Area Estimation with Applications to Poverty Measurement 85
Jan Pablo Burgard, Ralf Münnich and Thomas Zimmermann 5.1 Introduction 85 5.2 Sampling Designs in our Study 87 5.3 Estimation of Poverty Indicators 90 5.3.1 Design-based Approaches 90 5.3.2 Model-based Estimators 92 5.4 Monte Carlo Comparison of Estimation Methods and Designs 96 5.5 Summary and Outlook 105 Acknowledgements 106 References 106 6 Model-assisted Methods for Small Area Estimation of Poverty Indicators 109
Risto Lehtonen and Ari Veijanen 6.1 Introduction 109 6.1.1 General 109 6.1.2 Concepts and Notation 110 6.2 Design-based Estimation of Gini Index for Domains 111 6.2.1 Estimators 111 6.2.2 Simulation Experiments 114 6.2.3 Empirical Application 116 6.3 Model-assisted Estimation of At-risk-of Poverty Rate 117 6.3.1 Assisting Models in GREG and Model Calibration 117 6.3.2 Generalized Regression Estimation 119 6.3.3 Model Calibration Estimation 120 6.3.4 Simulation Experiments 122 6.3.5 Empirical Example 123 6.4 Discussion 124 6.4.1 Empirical Results 124 6.4.2 Inferential Framework 125 6.4.3 Data Infrastructure 125 References 126 7 Variance Estimation for Cumulative and Longitudinal Poverty Indicators from Panel Data at Regional Level 129
Gianni Betti, Francesca Gagliardi and Vijay Verma 7.1 Introduction 129 7.2 Cumulative vs. Longitudinal Measures of Poverty 130 7.2.1 Cumulative Measures 130 7.2.2 Longitudinal Measures 131 7.3 Principle Methods for Cross-sectional Variance Estimation 131 7.4 Extension to Cumulation and Longitudinal Measures 133 7.5 Practical Aspects: Specification of Sample Structure Variables 134 7.6 Practical Aspects: Design Effects and Correlation 136 7.6.1 Components of the Design Effect 136 7.6.2 Estimating the Components of Design Effect 138 7.6.3 Estimating other Components using Random Grouping of Elements 139 7.7 Cumulative Measures and Measures of Net Change 140 7.7.1 Estimation of the Measures 140 7.7.2 Variance Estimation 141 7.8 An Application to Three Years' Averages 141 7.8.1 Computation Given Limited Information on Sample Structure in EU-SILC 141 7.8.2 Direct Computation 144 7.8.3 Empirical Results 145 7.9 Concluding Remarks 146 References 147 Part III SMALL AREA ESTIMATION MODELING AND ROBUSTNESS 8 Models in Small Area Estimation when Covariates are Measured with Error 151
Serena Arima, Gauri S. Datta and Brunero Liseo 8.1 Introduction 151 8.2 Functional Measurement Error Approach for Area-level Models 153 8.2.1 Frequentist Method for Functional Measurement Error Models 154 8.2.2 Bayesian Method for Functional Measurement Error Models 156 8.3 Small Area Prediction with a Unit-level Model when an Auxiliary Variable is Measured with Error 156 8.3.1 Functional Measurement Error Approach for Unit-level Models 157 8.3.2 Structural Measurement Error Approach for Unit-level Models 160 8.4 Data Analysis 162 8.4.1 Example 1: Median Income Data 162 8.4.2 Example 2: SAIPE Data 165 8.5 Discussion and Possible Extensions 169 Acknowledgements 169 Disclaimer 170 References 170 9 Robust Domain Estimation of Income-based Inequality Indicators 171
Nikos Tzavidis and Stefano Marchetti 9.1 Introduction 171 9.2 Definition of Income-based Inequality Measures 172 9.3 Robust Small Area Estimation of Inequality Measures with M-quantile Regression 173 9.4 Mean Squared Error Estimation 176 9.5 Empirical Evaluations 176 9.6 Estimating the Gini Coefficient and the Quintile Share Ratio for Unplanned Domains in Tuscany 180 9.7 Conclusions 183 References 185 10 Nonparametric Regression Methods for Small Area Estimation 187
M. Giovanna Ranalli, F. Jay Breidt and Jean D. Opsomer 10.1 Introduction 187 10.2 Nonparametric Methods in Small Area Estimation 188 10.2.1 Nested Error Nonparametric Unit Level Model Using Penalized Splines 189 10.2.2 Nested Error Nonparametric Unit Level Model Using Kernel Methods 191 10.2.3 Generalized Responses 192 10.2.4 Robust Approaches 192 10.3 A Comparison for the Estimation of the Household Per-capita Consumption Expenditure in Albania 195 10.4 Concluding Remarks 202 References 202 Part IV SPATIO-TEMPORAL MODELING OF POVERTY 11 Area-level Spatio-temporal Small Area Estimation Models 207
María Dolores Esteban, Domingo Morales and Agustín Pérez 11.1 Introduction 207 11.2 Extensions of the Fay–Herriot Model 209 11.3 An Area-level Model with MA(1) Time Correlation 212 11.4 EBLUP and MSE 214 11.5 EBLUP of Poverty Proportions 215 11.6 Simulations 216 11.6.1 Simulation 1 216 11.6.2 Simulation 2 217 11.7 R Codes 220 11.8 Concluding Remarks 220 Appendix 11.A: MSE Components 221 11.A.1 Calculation of g1(
A comprehensive guide to implementing SAE methods for poverty studies and poverty mapping There is an increasingly urgent demand for poverty and living conditions data, in relation to local areas and/or subpopulations. Policy makers and stakeholders need indicators and maps of poverty and living conditions in order to formulate and implement policies, (re)distribute resources, and measure the effect of local policy actions. Small Area Estimation (SAE) plays a crucial role in producing statistically sound estimates for poverty mapping. This book offers a comprehensive source of information regarding the use of SAE methods adapted to these distinctive features of poverty data derived from surveys and administrative archives. The book covers the definition of poverty indicators, data collection and integration methods, the impact of sampling design, weighting and variance estimation, the issue of SAE modelling and robustness, the spatio-temporal modelling of poverty, and the SAE of the distribution function of income and inequalities.