12 December 2022
3:00–6:00 PM CET
Model-based small area estimation approaches are in great demand by institutions interested in reliable statistics at disaggregated levels. Such tools overcome estimation challenges using small or incomplete survey data, in part, by combining data from multiple sources. The course starts with an overview of important small area estimation concepts. Area-level models are the main focus of the course, as proven to be of great practical use. Hierarchical Bayes inference is adopted. Software programs are provided for the model fit, validation, prediction, and comparison, with model specification scrips in R STAN. Illustrations use public-use data from the U.S. Census Bureau, for state-level poverty rate estimation research.
Learning outcomes to be covered
Learn how to conduct area-level model-based small area estimation analyses using freely available software.
Description of course materials for online teaching
Materials will be shared with the participants in the form of PDF slides and R programs.
Proposed delivery structure, including elements of engagement
The participants are encouraged to install R and STAN, prior to the course. They will be encouraged to run the analyses during the course. Exercises will be provided at the end of the course.
The course is intended for statisticians interested in hierarchical Bayes model-based small area estimation.
Knowledge of mathematical statistics and a course in linear regression and mixed models would be helpful. Previous exposure to hierarchical models, Bayesian inference, and R statistical language are an asset, but are not required.
Level of instruction
· Rao, J. N. K., & Molina, I. (2015). Small Area Estimation. Second edi. Hoboken, New.
· Erciulescu, A.L. (2019). Hierarchical models in the production of official statistics: a discussion of some practical aspects. Available here.
· Erciulescu A.L., Franco C., Lahiri P. (2021) “The Use of Administrative Data in Small Area Estimation” Administrative Records for Survey Methodology. Ed. A. Y. Chun, Ed. M. Larsen, Co-Ed. J. Reiter, Co-Ed. G. Durrant. Wiley