报告题目：Semiparametric inference for inequality measures under nonignorable nonresponse using callback data

报告人：王淳林 副教授

报告时间：2025年12月27日15:30-18:30

报告地点：数学与统计学院210会议室

邀请人：熊贤祝

邀请单位：福州大学数学与统计学院

报告摘要：This paper develops semiparametric methods for the estimation and inference of widely used inequality measures when survey data are subject to nonignorable nonresponse, a challenging setting in which response probabilities depend on the unobserved income values. Nonignorable nonresponse is common in household income surveys and renders many standard inference procedures invalid due to the lack of population-representative data. We address this problem by exploiting callback data, which provide auxiliary information from repeated contact attempts, and then by adopting a semiparametric response model that leaves the income distribution unspecified. We construct semiparametric full-likelihood estimators of the income distribution and the associated inequality measures, and further derive their large-sample properties, including explicit variance expressions for a broad class of functionals. To facilitate computation, we propose a stable and computationally convenient expectation–maximization algorithm, whose steps involve closed-form expressions or reduce to fitting a standard logistic regression. Simulation evidence shows that the proposed procedures effectively correct nonresponse bias and achieve efficient inference across a range of income distributions. An application to Consumer Expenditure Survey data illustrates the practical gains from incorporating callback information on inference of inequality measures.

报告人简介：王淳林，现任厦门大学经济学院统计学与数据科学系与王亚南经济研究院副教授，博士生导师。加拿大滑铁卢大学统计学博士。目前研究方向包括：基于经验似然和密度比模型的半参数与非参数统计推断方法，及其在经济社会和生物医学中的应用。主持国家自然科学基金、教育部人文社会科学项目等多项研究课题，在Statistica Sinica, JMVA, CSDA, Statistics in Medicine等统计学期刊发表论文数篇。曾获厦门大学第十三届英语教学比赛特等奖、最佳教学设计奖。