Description
Book SynopsisA practical, one-stop reference on the theory and applications of statistical data editing and imputation techniques Collected survey data are vulnerable to error. In particular, the data collection stage is a potential source of errors and missing values.
Table of ContentsPreface ix
1 Introduction to Statistical Data Editing and Imputation 1
1.1 Introduction 1
1.2 Statistical Data Editing and Imputation in the Statistical Process 4
1.3 Data, Errors, Missing Data, and Edits 6
1.4 Basic Methods for Statistical Data Editing and Imputation 13
1.5 An Edit and Imputation Strategy 17
References 21
2 Methods for Deductive Correction 23
2.1 Introduction 23
2.2 Theory and Applications 24
2.3 Examples 27
2.4 Summary 55
References 55
3 Automatic Editing of Continuous Data 57
3.1 Introduction 57
3.2 Automatic Error Localization of Random Errors 59
3.3 Aspects of the Fellegi–Holt Paradigm 63
3.4 Algorithms Based on the Fellegi–Holt Paradigm 65
3.5 Summary 101
3.A Appendix: Chernikova’s Algorithm 103
References 104
4 Automatic Editing: Extensions to Categorical Data 111
4.1 Introduction 111
4.2 The Error Localization Problem for Mixed Data 112
4.3 The Fellegi–Holt Approach 115
4.4 A Branch-and-Bound Algorithm for Automatic Editing of Mixed Data 129
4.5 The Nearest-Neighbor Imputation Methodology 140
References 158
5 Automatic Editing: Extensions to Integer Data 161
5.1 Introduction 161
5.2 An Illustration of the Error Localization Problem for Integer Data 162
5.3 Fourier–Motzkin Elimination in Integer Data 163
5.4 Error Localization in Categorical, Continuous, and Integer Data 172
5.5 A Heuristic Procedure 182
5.6 Computational Results 183
5.7 Discussion 187
References 189
6 Selective Editing 191
6.1 Introduction 191
6.2 Historical Notes 193
6.3 Micro-selection: The Score Function Approach 195
6.4 Selection at the Macro-level 208
6.5 Interactive Editing 212
6.6 Summary and Conclusions 217
References 219
7 Imputation 223
7.1 Introduction 223
7.2 General Issues in Applying Imputation Methods 226
7.3 Regression Imputation 230
7.4 Ratio Imputation 244
7.5 (Group) Mean Imputation 246
7.6 Hot Deck Donor Imputation 249
7.7 A General Imputation Model 255
7.8 Imputation of Longitudinal Data 261
7.9 Approaches to Variance Estimation with Imputed Data 264
7.10 Fractional Imputation 271
References 272
8 Multivariate Imputation 277
8.1 Introduction 277
8.2 Multivariate Imputation Models 280
8.3 Maximum Likelihood Estimation in the Presence of Missing Data 285
8.4 Example: The Public Libraries 295
References 297
9 Imputation Under Edit Constraints 299
9.1 Introduction 299
9.2 Deductive Imputation 301
9.3 The Ratio Hot Deck Method 311
9.4 Imputing from a Dirichlet Distribution 313
9.5 Imputing from a Singular Normal Distribution 318
9.6 An Imputation Approach Based on Fourier–Motzkin Elimination 334
9.7 A Sequential Regression Approach 338
9.8 Calibrated Imputation of Numerical Data Under Linear Edit Restrictions 343
9.9 Calibrated Hot Deck Imputation Subject to Edit Restrictions 349
References 358
10 Adjustment of Imputed Data 361
10.1 Introduction 361
10.2 Adjustment of Numerical Variables 362
10.3 Adjustment of Mixed Continuous and Categorical Data 377
References 389
11 Practical Applications 391
11.1 Introduction 391
11.2 Automatic Editing of Environmental Costs 391
11.3 The EUREDIT Project: An Evaluation Study 400
11.4 Selective Editing in the Dutch Agricultural Census 420
References 426
Index 429
