Recovering Income Distribution in the Presence of Interval-Censored Data

This paper proposes a method to analyze interval-censored data, using multiple imputationbased on a heteroskedastic interval regression approach. The proposed model aims to obtain a synthetic data set that canbe used for standard analysis, including standard linear regression, quantile regression, or poverty and inequalityestimation. The paper presents two applications to show the performance of the method. First, it runs a Monte Carlosimulation to show the method's performance under the assumption of multiplicative heteroskedasticity, with andwithout conditional normality. Second, it uses the proposed methodology to analyze labor income data in Grenada for2013–20, where the salary data are interval-censored according to the salary intervals prespecified in the surveyquestionnaire. The results obtained are consistent across both exercises.