Publications Repository

Publications Repository

View item 
  •   Home
  • ICR Divisions
  • Other ICR Research
  • View item
  • Home
  • ICR Divisions
  • Other ICR Research
  • View item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Bayesian disclosure risk assessment: predicting small frequencies in contingency tables

Thumbnail
Date
2007-11
ICR Author
Webb, Emily
Type
Journal Article
Metadata
Show full item record
Abstract
We propose an approach for assessing the risk of individual identification in the release of categorical data. This requires the accurate calculation of predictive probabilities for those cells in a contingency table which have small sample frequencies, making the problem somewhat different from usual contingency table estimation, where interest is generally focused on regions of high probability. Our approach is Bayesian and provides posterior predictive probabilities of identification risk. By incorporating model uncertainty in our analysis, we can provide more realistic estimates of disclosure risk for individual cell counts than are provided by methods which ignore the multivariate structure of the data set.
URI
https://repository.icr.ac.uk/handle/internal/1869
Collections
  • Other ICR Research
Subject
categorical data; identification; model uncertainty; prediction GRAPHICAL MODELS; MICRODATA
Language
eng
License start date
2007-11
Citation
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES C-APPLIED STATISTICS, 2007, 56 (5), pp. 551 - 570

Browse

All of ICR repositoryICR DivisionsBy issue dateAuthorsTitlesPublication TypesThis collectionBy issue dateAuthorsTitlesPublication Types
  • Login
  • Registered office: The Institute of Cancer Research, 123 Old Brompton Road, London, SW7 3RP
    A Charity, Not for Profit. Company Limited by Guarantee.
    Registered in England No. 534147. VAT Registration No. GB 849 0581 02.