Analysing Sensitive Data from Dynamically-Generated Overlapping Contingency Tables

Abstract Contingency tables provide a convenient format to publish summary data from confidential survey and administrative records that capture a wide range of social and economic information. By their nature, contingency tables enable aggregation of potentially sensitive data, limiting disclosure of identifying information. Furthermore, censoring or perturbation can be used to desensitise low cell counts when they arise. However, access to detailed cross-classified tables for research is often restricted by data custodians when too many censored or perturbed cells are required to preserve privacy. In this article, we describe a framework for selecting and combining log-linear models when accessible data is restricted to overlapping marginal contingency tables. The approach is demonstrated through application to housing transition data from the Australian Census Longitudinal Data set provided by the Australian Bureau of Statistics.

• Publication year

  2020

• Venue

  Journal of Official Statistics

• Publication date

  2020-06-01

• Fields of study

  Not labeled

• Identifiers
  DOI 10.2478/jos-2020-0015
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Graphical Models

  2020influential reference
• Thinning a Triangulation of a Bayesian Network or Undirected Graph to Create a Minimal Triangulation

  2017cited by this paper
• Sufficiency Revisited: Rethinking Statistical Algorithms in the Big Data Era

  2017cited by this paper
• The Australian Census Longitudinal Dataset: using record linkage to create a longitudinal sample from a series of cross‐sections

  2017cited by this paper
• The Australian Bureau of Statistics and releasing frequency tables via a remote server

  2016influential reference
• R: A language and environment for statistical computing.

  2014cited by this paper
• Holistic housing pathways for Australian families through the childbearing years

  2014cited by this paper
• Statistical Confidentiality: Principles and Practice

  2011cited by this paper
• Decomposition and Model Selection for Large Contingency Tables

  2009influential reference
• Marginal models for dependent, clustered, and longitudinal categorical data

  2009influential reference
• Categorical Data Analysis

  2003cited by this paper
• Marginal models for categorical data

  2002cited by this paper
• Maximal Prime Subgraph Decomposition of Bayesian Networks

  2001cited by this paper
• Resampling for statistical confidentiality in contingency tables

  1999cited by this paper
• Regression Analysis of Count Data

  1998cited by this paper
• On the comparison of multinomial and Poisson log-linear models

  1996cited by this paper
• Optimal decomposition by clique separators

  1993cited by this paper
• Marginalization and Collapsibility in Graphical Interaction Models

  1990cited by this paper
• Decomposition of maximum likelihood in mixed graphical interaction models

  1989cited by this paper
• Generalized Linear Models

  1984cited by this paper
• Measures of Nominal-Ordinal Association

  1981cited by this paper
• ANALYZING COLLAPSED CONTINGENCY TABLES WITHOUT ACTUALLY COLLAPSING

  1980influential reference
• Markov Fields and Log-Linear Interaction Models for Contingency Tables

  1980cited by this paper
• Algorithmic Aspects of Vertex Elimination on Graphs

  1976cited by this paper
• Algorithmic aspects of vertex elimination

  1975cited by this paper
• Discrete Multivariate Analysis: Theory and Practice

  1975influential reference
• A new look at the statistical model identification

  1974cited by this paper
• GENERALIZED LINEAR MODELS

  1972cited by this paper
• Maximum Likelihood in Three-Way Contingency Tables

  1963cited by this paper

• Marginal Models: an Overview

  2023influential citation