Journal of Official Statistics, Vol.14, No.4, 1998. pp. 485–502
Disclosure Limitation Using Perturbation and Related Methods for Categorical Data
Stephen E. Fienberg, Udi E. Makov, and Russell J. Steele
Abstract:During the past twenty-five years, the field of disclosure protection has undergone a "statistical transformation' and has begun to utilize the advances that have occurred within the field of statistics itself as well as in a variety of areas of application. This article reexamines some of the approaches currently employed in statistical disclosure limitation methodology for categorical data, e.g., cell suppression and data swapping, and relates them to the more conventional statistical methods associated with loglinear models and the simulation of exact distributions. It ties this perturbation approach to a general framework for the use of simulated data which we described earlier in Fienberg (1996) and Fienberg, Steele, and Makov (1996).
Keywords:Bootstrap; cell suppression; confidentiality; contingency table analysis; data swapping; loglinear models; multiple imputation.
Copyright © Statistics Sweden 1996-2018. Open AccessISSN 0282-423XCreated and Maintained by OKS Group