de Waal / Pannekoek / Scholtus Handbook of Statistical Data Editing and Imputation

1. Auflage 2011
ISBN: 978-0-470-90483-1
Verlag: John Wiley & Sons
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 464 Seiten, E-Book

Reihe: Wiley Handbooks in Survey Methodology

A practical, one-stop reference on the theory and applications ofstatistical data editing and imputation techniques
Collected survey data are vulnerable to error. In particular,the data collection stage is a potential source of errors andmissing values. As a result, the important role of statistical dataediting, and the amount of resources involved, has motivatedconsiderable research efforts to enhance the efficiency andeffectiveness of this process. Handbook of Statistical Data Editingand Imputation equips readers with the essential statisticalprocedures for detecting and correcting inconsistencies and fillingin missing values with estimates. The authors supply an easilyaccessible treatment of the existing methodology in this field,featuring an overview of common errors encountered in practice andtechniques for resolving these issues.
The book begins with an overview of methods and strategies forstatistical data editing and imputation. Subsequent chaptersprovide detailed treatment of the central theoretical methods andmodern applications, with topics of coverage including:
* Localization of errors in continuous data, with an outline ofselective editing strategies, automatic editing for systematic andrandom errors, and other relevant state-of-the-art methods
* Extensions of automatic editing to categorical data and integerdata
* The basic framework for imputation, with a breakdown of keymethods and models and a comparison of imputation with theweighting approach to correct for missing values
* More advanced imputation methods, including imputation underedit restraints
Throughout the book, the treatment of each topic is presented ina uniform fashion. Following an introduction, each chapter presentsthe key theories and formulas underlying the topic and thenillustrates common applications. The discussion concludes with asummary of the main concepts and a real-world example thatincorporates realistic data along with professional insight intocommon challenges and best practices.
Handbook of Statistical Data Editing and Imputation is anessential reference for survey researchers working in the fields ofbusiness, economics, government, and the social sciences whogather, analyze, and draw results from data. It is also a suitablesupplement for courses on survey methods at the upper-undergraduateand graduate levels.

Autoren/Hrsg.

de Waal, Ton

Pannekoek, Jeroen

Scholtus, Sander

Inhaltsverzeichnis

Preface.
1 Introduction to statistical data editing andimputation.
1.1 Introduction.
1.2 Statistical data editing and imputation in the statisticalprocess.
1.3 Data, errors, missing data and edits.
1.4 Basic methods for statistical data editing andimputation.
1.5 An edit and imputation strategy.
2 Methods for deductive correction.
2.1 Introduction.
2.2 Theory and applications.
2.3 Examples.
2.4 Summary.
3 Automatic editing of continuous data.
3.1 Introduction.
3.2 Automatic error localisation of random errors.
3.3 Aspects of the Fellegi-Holt paradigm.
3.4 Algorithms based on the Fellegi-Holt paradigm.
3.5 Summary.
4 Automatic editing: extensions to categorical data.
4.1 Introduction.
4.2 The error localisation problem for mixed data.
4.3 The Fellegi-Holt approach.
4.4 A branch-and-bound algorithm for automatic editing of mixeddata.
4.5 The Nearest-neighbour Imputation Methodology.
5 Automatic editing: extensions to integer data.
5.1 Introduction.
5.2 An illustration of the error localisation problem forinteger data.
5.3 Fourier-Motzkin elimination in integer data.
5.4 Error localisation in categorical, continuous and integerdata.
5.5 A heuristic procedure.
5.6 Computational results.
5.7 Discussion.
6 Selective editing.
6.1 Introduction.
6.2 Historical notes.
6.3 Micro-selection: the score function approach.
6.4 Selection at macro-level.
6.5 Interactive editing.
6.6 Summary and conclusions.
7 Imputation.
7.1 Introduction.
7.2 General issues in applying imputation methods.
7.3 Regression imputation.
7.4 Ratio imputation.
7.5 (Group) mean imputation.
7.6 Hot deck donor imputation.
7.7 A general imputation model.
7.8 Imputation of longitudinal data.
7.9 Approaches to variance estimation with imputed data.
7.10 Fractional imputation.
8 Multivariate imputation.
8.1 Introduction.
8.2 Multivariate imputation models.
8.3 Maximum likelihood estimation in the presence of missingdata.
8.4 Example: the public libraries.
9 Imputation under edit constraints.
9.1 Introduction.
9.2 Deductive imputation.
9.3 The ratio hot deck method.
9.4 Imputing from a Dirichlet distribution.
9.5 Imputing from a singular normal distribution.
9.6 An imputation approach based on Fourier-Motzkinelimination.
9.7 A sequential regression approach.
9.8 Calibrated imputation of numerical data under linear editrestrictions.
9.9 Calibrated hot deck imputation subject to editrestrictions.
10 Adjustment of imputed data.
10.1 Introduction.
10.2 Adjustment of numerical variables.
10.3 Adjustment of mixed continuous and categorical data.
11 Practical applications.
11.1 Introduction.
11.2 Automatic editing of environmental costs.
11.3 The EUREDIT project: an evaluation study.
11.4 Selective editing in the Dutch Agricultural Census.
Index.

Über Autor(innen)

Ton De Waal, PhD, is Head of the Department of Methodology atStatistics Netherlands, where he has also worked at the Division ofBusiness Statistics. Dr. de Waal has written numerous papers in hisareas of research interest, which include statistical data editingand imputation for business surveys and statistical disclosurecontrol.
Jeroen Pannekoek, PhD, is Senior Researcher in the Department ofMethodology at Statistics Netherlands, where he currently leads theresearch program on data processing methodologies. He has publishedseveral papers on discrete data models, measurement errors,interviewer effects, and disclosure control methods.
Sander Scholtus, MSc, is Researcher in the Department ofMethodology at Statistics Netherlands. He has conducted extensiveresearch on heuristic methods and algorithms for detecting andcorrecting errors in survey data.