On the discrepancy measures for the optimal equal probability partitioning in bayesian multivariate micro-aggregation

Kokolakis, G; Fouskakis, D

dc.contributor.author	Kokolakis, G	en
dc.contributor.author	Fouskakis, D	en
dc.date.accessioned	2014-03-01T01:28:55Z
dc.date.available	2014-03-01T01:28:55Z
dc.date.issued	2008	en
dc.identifier.issn	0176-4268	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/19034
dc.subject	Asymptotic theory	en
dc.subject	Convex partition	en
dc.subject	Discrepancy measures	en
dc.subject	Identity protection	en
dc.subject	Multivariate micro-aggregation	en
dc.subject	Probabilistic distances	en
dc.subject.classification	Mathematics, Interdisciplinary Applications	en
dc.subject.classification	Psychology, Mathematical	en
dc.subject.other	DISCLOSURE	en
dc.title	On the discrepancy measures for the optimal equal probability partitioning in bayesian multivariate micro-aggregation	en
heal.type	journalArticle	en
heal.identifier.primary	10.1007/s00357-008-9014-8	en
heal.identifier.secondary	http://dx.doi.org/10.1007/s00357-008-9014-8	en
heal.language	English	en
heal.publicationDate	2008	en
heal.abstract	Data holders, such as statistical institutions and financial organizations, have a very serious and demanding task when producing data for official and public use. It's about controlling the risk of identity disclosure and protecting sensitive information when they communicate data-sets among themselves, to governmental agencies and to the public. One of the techniques applied is that of micro-aggregation. In a Bayesian setting, micro-aggregation can be viewed as the optimal partitioning of the original data-set based on the minimization of an appropriate measure of discrepancy, or distance, between two posterior distributions, one of which is conditional on the original data-set and the other conditional on the aggregated data-set. Assuming d-variate normal data-sets and using several measures of discrepancy, it is shown that the asymptotically optimal equal probability m-partition of ℝd, with m 1/d ε ℕ, is the convex one which is provided by hypercubes whose sides are formed by hyperplanes perpendicular to the canonical axes, no matter which discrepancy measure has been used. On the basis of the above result, a method that produces a sub-optimal partition with a very small computational cost is presented. © 2008 Springer Science+Business Media, LLC.	en
heal.publisher	SPRINGER	en
heal.journalName	Journal of Classification	en
dc.identifier.doi	10.1007/s00357-008-9014-8	en
dc.identifier.isi	ISI:000262413100007	en
dc.identifier.volume	25	en
dc.identifier.issue	2	en
dc.identifier.spage	209	en
dc.identifier.epage	224	en