Introduction to Small Area Estimation

Large-scale sample surveys are not designed to produce reliable estimates for small population domains, e.g., geographical areas or population groups. This is because area sample sizes may be “too small” to compute reliable direct estimates. Therefore, small area estimation methods, that borrow strength information from auxiliary data e.g., the Census or administrative data, can be used to produce reliable estimates. This course covers widely adopted small area estimation methods based on the direct and model-based estimation approach and it is structured in three parts. The first part is about the introduction to the small area estimation problem and the use of direct estimators to produce small area estimates. In the second part, we introduce the unit-level approach based on the Battese, Harter and Fuller model, assuming that auxiliary information is available at unit-level. The third part is on the area-level approach, based on the Fay-Herriot model. This approach is useful when the auxiliary information is available are area-level only. The course will also focus on evaluation of the small area estimates via diagnostics tools and simulation studies. Applications will also be provided in each part of the course.

Practical applications and examples will be in R. We will show some common R packages available to users and applications will be based both on real data and simulation studies.

Course leader

Angelo Moretti is an assistant professor in Statistics at Utrecht University in the Department of Methodology and Statistics.

Target group

You will find the course useful if:
- you are a social scientist or statistician at PhD-level or beyond, working on academic research projects, or a researcher working in the survey field.

Prerequisites:
- Introductory knowledge of regression models and statistical inference is assumed.
- Basic knowledge of R language is required.
- You should have a working installation of R and RStudio
- We ask you to send us a brief motivation letter (up to 200 words), in which you write a short list of your experience with surveys and statistical analyses. Also include in the motivation letter what courses related to survey design, estimation and analysis were taken. Please send the document as soon as you have registered to summerschool@gesis.org.

Course aim

By the end of the course, you will:
- understand the small area estimation problem,
- understand the appropriate small area estimation approach given a specific survey data,
- be able to apply and validate relevant small area estimation methods based on the area-level and unit-level approach,
- implement the methods in R software,
- present and visualize the results and analyses.

Credits info

4 EC
- Certificate of attendance issued upon completion.

Optional bookings:
The University of Mannheim acknowledges the workload for regular attendance, satisfactory work on daily assignments and for submitting a paper of 5000 words to the lecturer(s) by 15 October at the latest with 4 ECTS (70 EUR administration fee).

Fee info

EUR 550: Student/PhD student rate
EUR 825: Academic/non-profit rate

The rates include the tuition fee, course materials, the academic program, and coffee/tea breaks.

Scholarships

Scholarships are available from the European Survey Research Association (ESRA), see more information on our website