Special Issue "Small Area Estimation: Theories, Methods and Applications"

Dear Colleagues,

Small area estimation is important to many government agencies across the world as it is a vehicle used to build official statistics. It has grown to enormous importance over the past fifty years and is in practice today as the demand for small area statistics has significantly increased worldwide. Currently, there is a strong need for formulating policies and programs for the allocation of government funds and regional planning.

Recent advances in statistical methodology and computation, coupled with strong demand for disaggregated estimates, have led to the field of small area estimation expanding in many directions. Robust estimators and generalized linear mixed models accommodate non-normal distributions in unit-level responses. Non-parametric models allow for further flexibility in describing distributional forms and indirect relationships to covariates. Measurement error models and variable selection methods aid in the use of increasingly complex auxiliary data sources.  Spatio-temporal and multivariate models capture dependencies among areas and variables. Proper studies on nonresponse and selection bias guard against biased predictors when the selection mechanisms are related to the characteristic of interest. The bootstrap and the Bayesian paradigm facilitate the construction of accurate prediction intervals and mean square errors of estimators, even for complex models, where parameters are nonlinear functions of response variables and there are clustering effects. Developments in statistical methods have enabled many important applications that rely on small area estimates. With all these new methods, judicious pooling of small areas is essential. Non-probability sampling is an emerging area.

I would like this Special Issue to contain papers in small area estimation that have strong theories, methods, and applications, and a mix of these broad areas. Therefore, I welcome papers that develop innovative small area methods or that demonstrate sound applications of small area estimation to problems of practical interest. Some prominent researchers in small area estimation have already agreed to submit papers for possible publication. I look forward to receiving your submission.

Prof. Dr. Balgobin Nandram
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Stats is an international peer-reviewed open access quarterly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1200 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• correlated data
• nonlinear model model
• non-parametric model
• hierarchical and multi-level models
• nonresponse
• selection bias
• Bayesian methodology
• bootstrap and resampling
• spatio-temporal models

Title: Transformation-based variable selection for three-fold sub-subarea level models in small area estimation

Authors: Song Cai and J.N.K. Rao

Abstract: We consider surveys that are conducted in a hierarchical fashion, where a country is partitioned into census divisions, states are chosen randomly from these divisions, then counties are randomly selected from each chosen state, and finally survey units are selected from each chosen county. To estimate a county-level parameter, we propose a three-fold sub-subarea level model with a division-level random effect, a state-level within division random effect and a county within state random effect. To obtain a parsimonious model, we proposal two easy-to-implement transformation-based methods to select covariate variables. One transformation is model-parameter dependent, and the other transformation is model-parameter free. The effectiveness of the two variable selection methods will be assessed by a simulation study. Results of a simulation study on the relative performance of different variable selection methods are reported. Empirical best linear unbiased predictors (EBLUPs) of county means under the three-fold sub-subarea level models provide more efficient estimators for non-sampled counties within a sampled state as well as non-sampled counties within a non-sampled state, compared to traditional synthetic estimators.