Bayesian nonparametric disclosure risk estimation via mixed effects log-linear models

DOI10.1214/15-AOAS807zbMATH Open1454.62107arXiv1306.5995MaRDI QIDQ2349591FDOQ2349591

Authors: Cinzia Carota, Roberto Leombruni, Silvia Polettini, M. Filippone 

Publication date: 17 June 2015

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: Statistical agencies and other institutions collect data under the promise to protect the confidentiality of respondents. When releasing microdata samples, the risk that records can be identified must be assessed. To this aim, a widely adopted approach is to isolate categorical variables key to the identification and analyze multi-way contingency tables of such variables. Common disclosure risk measures focus on sample unique cells in these tables and adopt parametric log-linear models as the standard statistical tools for the problem. Such models often have to deal with large and extremely sparse tables that pose a number of challenges to risk estimation. This paper proposes to overcome these problems by studying nonparametric alternatives based on Dirichlet process random effects. The main finding is that the inclusion of such random effects allows us to reduce considerably the number of fixed effects required to achieve reliable risk estimates. This is studied on applications to real data, suggesting, in particular, that our mixed models with main effects only produce roughly equivalent estimates compared to the all two-way interactions models, and are effective in defusing potential shortcomings of traditional log-linear models. This paper adopts a fully Bayesian approach that accounts for all sources of uncertainty, including that about the population frequencies, and supplies unconditional (posterior) variances and credible intervals.

Full work available at URL: https://arxiv.org/abs/1306.5995

Recommendations

• Bayesian nonparametric disclosure risk assessment
• Estimating identification disclosure risk using mixed membership models
• A Measure of Disclosure Risk for Microdata
• Optimal disclosure risk assessment
• Some issues concerning disclosure risk in contingency tables

zbMATH Keywords

log-linear modelsconfidentialitymixed effects modelsDirichlet processBayesian nonparametric modelsdisclosure risk

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Contingency tables (62H17)

Cites Work

• Inference from iterative simulation using multiple sequences
• Bayesian Density Estimation and Inference Using Mixtures
• Three centuries of categorical data analysis: Log-linear models and maximum likelihood estima\-tion
• A Bayesian analysis of some nonparametric problems
• Accurate Approximations for Posterior Moments and Marginal Densities
• Title not available (Why is that?)
• Maximum likelihood estimation in log-linear models
• 10.1162/jmlr.2003.3.4-5.993
• Assessing identification risk in survey microdata using log-linear models
• Title not available (Why is that?)
• Hierarchical Dirichlet Processes
• Ferguson distributions via Polya urn schemes
• Nonparametric hierarchical Bayes via sequential imputations
• On a class of Bayesian nonparametric estimates: I. Density estimates
• Describing disability through individual-level mixture models for multivariate binary data
• Modeling Unobserved Sources of Heterogeneity in Animal Abundance Using a Dirichlet Process Prior
• Discussion of the paper: ``Sampling schemes for generalized linear Dirichlet process random effects models by M. Kyung, J. Gill, and G. Casella
• Title not available (Why is that?)
• Estimating identification disclosure risk using mixed membership models

Cited In (8)

• Assessing identification risk in survey microdata using log-linear models
• A Measure of Disclosure Risk for Microdata
• Assessing Bayesian Semi‐Parametric Log‐Linear Models: An Application to Disclosure Risk Estimation
• Optimal disclosure risk assessment
• Estimating identification disclosure risk using mixed membership models
• A Bayesian semiparametric GLMM for historical and newly collected presence-only data: an application to species richness of Ross sea mollusca
• Bayesian nonparametric disclosure risk assessment
• Identification, data combination, and the risk of disclosure

This page was built for publication: Bayesian nonparametric disclosure risk estimation via mixed effects log-linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2349591)