Statistical confidentiality. Principles and practice. (Q610154)

This is an interesting book that should be read not only by statisticians but by every person involved with managing large amounts of data. If your data is sensitive it will be not only more interesting but useful. The book is aimed at people looking for protecting their statistical data bases. The oeuvre is concerned with the main aspects of statistical confidentiality in the identification of dilemmas that it poses to the owners of data bases and the accessibility to the data and protection they must guarantee but to the protection to the individuality of data providers, against the attack from people looking into the statistical data. A main aspect in this problem is characterized by the dialectic Statistical Disclosure Risks (SDRs) and Data Stewardship Organizations (DSO). A main issue is the reduction of disclosure risks, its limitations as well as the techniques allowing risk assessment in current use (risk and assessment measuring) and the identification of a lower bound of the risks. DSO should select a SDR in terms of the protection needed by their tables. Various methods for protecting source tables are discussed. Due to the particularities of the weakness of micro data masking of them (matrices and sampling masking, suppressing, aggregating, swapping, nosing, coding) is proposed and some experiences with their use are discussed. The book has a chapter devoted to presenting some particular SDRs motivated by masking methods as regression and game theory. The restriction to the access of the information is discussed. It is preferred by persons with access posing contradictions as an additional complication for the researchers motivating special challenges to the data owners. The final chapter considers the future of the previously considered issues. The challenges go from the dialectic of the signification of the pair privacy-confidentiality to the development of DSO and SDR strategies. The authors quote the importance of the problems that academics ought to study for aiding the research that governmental statisticians are doing. A glossary with more than 60 terms is given at the end of the book together with more than bibliographic 200 entries