IASS - Webinar 51: Inferential Issues in Model-based Small Area Estimation: Some History and Selective Recent Developments
| Date | 30 Apr 2025 |
| Time | 13:00 (GMT+02:00) - 14:30 (GMT+02:00) |
| Level of instruction | Intermediate |
| Instructor |
J.N.K. Rao
|
| Registration fee | |
Model-based small area estimation (SAE) has seen rapid growth over the past 25 years or so due to increasing demand for reliable small area estimators of totals, means and more complex parameters. Direct area-specific estimators are either unreliable (due to small area-specific sample sizes) or not feasible (due to zero area-specific sample sizes). It is therefore necessary to “borrow strength” through implicit or explicit models linking related areas. We focus on explicit area-level and unit-level models. After a brief historical sketch of SAE, we focus on selective recent developments (mostly after 2015). Under a basic area level model, we consider semi-parametric MSE estimation and robust estimation of small area means under model misspecification. Under a basic unit level model, methods for estimating complex small area parameters, such as poverty indicators, will be presented, including robust estimation in the presence of outliers. A comparison of area level and unit level models when unit level data are available will also be considered. Estimation methods studied included empirical best linear unbiased prediction (EBLUP), empirical best (EB) and hierarchical Bayes (HB).
Instructors
About the instructor
J.N.K. Rao is a Distinguished Research Professor at Carleton University, Ottawa. He is a leading authority in survey sampling theory and methods and empirical likelihood methods for survey data. In recent years, his research focused on small area estimation (SAE), a topic of great practical interest due to growing demand for
reliable local area statistics. His two Wiley books on this topic published in 2003 and 2015 are regarded as standard references on this topic and highly cited. In recognition of his pioneering contributions to this topic, he received the first Award for outstanding contributions to SAE at an international conference on SAE in Paris, 2017. He is also known for his pioneering 1968 paper on scale-load approach to non-parametric inference from survey data and 20 years later rediscovered as empirical likelihood. This topic has received much attention in recent years due to its ability to provide reliable nonparametric confidence intervals similar to likelihood ratio intervals in the parametric framework. In recognition of his outstanding contributions to survey sampling and other areas, Rao received many prestigious honours and awards: Fellow of the Royal Society of Canada (1991), American Statistical Association (1964) and Institute of Mathematical Statistics (1972). He received the Gold Medal from the Statistical Society of Canada (1993) and the Waksberg Award (2005).