Two-Phase Data Synthesis for Income: An Application to the NHIS

TL;DR

This paper introduces a two-phase Bayesian data synthesis method for generating synthetic income data, effectively handling skewness and zeros, and demonstrates its utility using NHIS data.

Contribution

The paper presents a novel two-phase synthesis approach for income data, improving upon single-phase methods by better modeling skewness and zero-inflation.

Findings

Two-phase synthesis outperforms single-phase in utility and risk profiles.

Bayesian models effectively handle skewed and zero-inflated income data.

Application to NHIS data demonstrates practical utility.

Abstract

We propose a two-phase synthesis process for synthesizing income, a sensitive variable which is usually highly-skewed and has a number of reported zeros. We consider two forms of a continuous income variable: a binary form, which is modeled and synthesized in phase 1; and a non-negative continuous form, which is modeled and synthesized in phase 2. Bayesian synthesis models are proposed for the two-phase synthesis process, and other synthesis models are readily implementable. We demonstrate our methods with applications to a sample from the National Health Interview Survey (NHIS). Utility and risk profiles of generated synthetic datasets are evaluated and compared to results from a single-phase synthesis process.