Accounting for Dependence in Small Area Estimation

Chair: Alan Welsh

International demand for high quality, timely, small-domain data from official statistical agencies is steadily increasing and these small domain estimates can be extremely uncertain. Consequently, there is an expanding need for innovative methods that increase the precision and level of usability in publicly available data. In addition, enabling data users to obtain stable estimates for under-sampled or un-sampled geographies and demographies based on data that are not publicly available substantially enhances national and international policy and scientific impact. In this direction, spatial and spatio-temporal, hierarchical statistical models are effective in increasing the statistical stability of small domain estimates while maintaining an acceptable degree of domain-specific focus. This session highlights recent advances in hierarchical statistical approaches to achieving these goals. The approaches are broadly applicable to official statistics worldwide and to other contexts such as agricultural and epidemiology, among others.


Small Area Estimation for High-Dimensional Multivariate Spatio-Temporal Count Data

Small area estimation of count data has become a research topic of widespread interest due to the ever-increasing need to produce more precise estimates for undersampled/unsampled geographies. This problem becomes more exacerbated when one acknowledges that many data sources also report related variables of interest that are referenced at different levels of spatial aggregation and […]

Quantifying and Mitigating Spatial Aggregation Error

The modifiable areal unit problem and the ecological fallacy are known problems that occur when modeling multiscale spatial processes. We investigate how these forms of spatial aggregation error can be mitigated and guide a regionalization over a spatial domain of interest. By “regionalization” we mean a specification of geographies that define the spatial support for […]

This session was organised by Scott H. Holan.