Bayesian Marked Point Process Modeling for Generating Fully Synthetic Public Use Data with Point-Referenced Geography

TL;DR

This paper introduces a Bayesian marked point process model to generate fully synthetic spatial data, preserving statistical and spatial properties while protecting confidentiality, offering an alternative to traditional data redaction methods.

Contribution

The paper presents a novel Bayesian marked point process approach for creating fully synthetic spatial data that maintains data utility and confidentiality.

Findings

Successfully modeled mortality data from Durham, North Carolina.

Generated synthetic data preserving spatial dependence.

Outperformed traditional coarsening and perturbation methods.

Abstract

Many data stewards collect confidential data that include fine geography. When sharing these data with others, data stewards strive to disseminate data that are informative for a wide range of spatial and non-spatial analyses while simultaneously protecting the confidentiality of data subjects' identities and attributes. Typically, data stewards meet this challenge by coarsening the resolution of the released geography and, as needed, perturbing the confidential attributes. When done with high intensity, these redaction strategies can result in released data with poor analytic quality. We propose an alternative dissemination approach based on fully synthetic data. We generate data using marked point process models that can maintain both the statistical properties and the spatial dependence structure of the confidential data. We illustrate the approach using data consisting of mortality records from Durham, North Carolina.