CMStatistics 2019: Start Registration
View Submission - CMStatistics
B0642
Title: Regularized multi-level models for small area estimation using both unit- and area-level data Authors:  Joscha Krause - Trier University (Germany) [presenting]
Jan Pablo Burgard - Trier University (Germany)
Ralf Muennich - University of Trier (Germany)
Abstract: The joint usage of unit- and area-level data for model-based small area estimation is investigated. The combination of levels within a single model encloses a variety of methodological problems. Firstly, it implies a critical decrease in degrees of freedom due to more model parameters that need to be estimated. This may destabilize model predictions in the presence of small samples. Secondly, unit- and area- level data has different distributional characteristics in terms of dispersion patterns and correlation structure. Thirdly, unit- and area- level data is usually subject to different kinds of measurement errors. We propose a multi-level model with level-specific regularizations to overcome these issues and use unit- and area-level data jointly for model-based small area estimation. In the process, we evaluate several mixed-norm regularizations to determine an optimal penalization strategy for a set of potential data constellations. All developed methods are tested within a Monte Carlo simulation study. Further, an empirical application is provided on the example of regional health measurement in Germany. We combine health survey data on the unit-level and aggregated micro census records on the area-level to estimate hypertension prevalence.