Tutorial: Small Area Estimation with R

Virgilio Gómez-Rubio, Imperial College, London, UK.

URL with more specific information



The aim of the workshop is to provide a practical introduction to Small Area Estimation (SAE) with R. At the moment, there is little training in this subject using R.

The workshop will start with an introduction to the analysis of survey data with design-based estimators, using packages 'survey' and 'sampling'. This will be followed by some other model-based estimators, such as linear regression and mixed-effects models using package 'nlme'. Package 'SAE', which is available at http://www.bias-project.org.uk/software, will also be employed for the computation of different EBLUP and Spatial EBLUP estimators. Bayesian models will be employed to illustrate the case of hierarchical (multilevel) models in Small Area Estimation. Furthermore, special attention will be paid to models that combine individual and aggregate information to produce estimates in areas where survey data are not available.

Whenever possible, examples will use real data sets. However, given that access to individual level survey data may be difficult to obtain due to confidentiality issues some examples will be developed using a simulated data set.

Draft programme

Workshop materials

The main packages used in this workshop are available in the Tasks Views 'Spatial' and 'SocialSciences'. Bayesian estimators will be computed using the freely-available WinBUGS, which will be called from R using package R2WinBUGS. The SAE package, and the needed WinBUGS code, are available from the BIAS Project web site at http://www.bias-project.org.uk. A web page on the course will be set in the former web site with all the course materials, including course slides, additional software and instructions on how to installed the required R packages.

Preliminary materials covering several parts of the course are already available in the vignette that is included in the 'SAE' package.


  1. J. C. Pinheiro and D. M. Bates (2000). Mixed-Effects Models in S and S-PLUS. Springer Verlag.
  2. J.N.K. Rao (2005). Small Area Estimation. Wiley & Sons, Inc.