2019/03/31 by Ray Chambers, Chambers, Ray, Enrico Fabrizi +3 · 1 citation
Decision Sciences · Mathematics · #Census and Population Estimation #Data Quality and Management #FOS: Computer and information sciences #Methodology (stat.ME) #Statistical Methods and Bayesian Inference #stat.ME
paper · pdf · doi:10.48550/arxiv.1904.00364
29 pages, 3 Figures
arxiv created 2019/03/31 · openalex publication_date 2019/03/31 · arxiv updated 2019/04/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In Small Area Estimation data linkage can be used to combine values of the variableof interest from a national survey with values of auxiliary variables obtained from another source like a population register. Linkage errors can induce bias when fitting regression models; moreover, they can create non-representative outliers in the linked data in addition to the presence of potential representative outliers. In this paper we adopt a secondary analyst's point view, assuming limited information is available on the linkage process, and we develop small area estimators based on linear mixed and linear M-quantile models to accommodate linked data containing a mix of both types of outliers. We illustrate the properties of these small area estimators, as well as estimators of their mean squared error, by means of model-based and design-based simulation experiments. These experiments show that the proposed predictors can lead to more efficient estimators when there is linkage error. Furthermore, the proposed mean-squared error estimation methods appear to perform well.