Online Program Home
My Program

Abstract Details

Activity Number: 47 - Statistical Analysis of Linked Data
Type: Invited
Date/Time: Sunday, July 29, 2018 : 4:00 PM to 5:50 PM
Sponsor: Survey Research Methods Section
Abstract #326553 Presentation
Title: Outlier Robust Inference Using Probabilistically Linked Data
Author(s): Nicola Salvati and Suojin Wang and Enrico Fabrizi and Raymond Chambers*
Companies: University of Pisa and Texas A&M University and Catholic University of Sacro Cuore and University of Wollongong
Keywords: Linkage Errors; Estimating Equations; Bias Correction; Secondary Analysis; Regression Analysis

Linkage errors occur when probability-based methods are used to link or match records from two or more distinct data sets corresponding to the same target population. These errors can lead to biased analytical decisions when they are ignored. We investigate an estimating equations approach to specifying bias corrected secondary analysis of probabilistically linked data, based on a realistic scenario of dependent linkage errors in a linear regression setting. We then develop outlier robust solutions when population auxiliary information in the form of population summary statistics is available. Our simulation results show that standard outlier robust methods under an incorrect assumption of independent linkage errors can lead to insufficient linkage error bias correction, while an outlier robust approach that allows for correlated linkage errors appears to substantially correct this bias.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program