We present a Bayesian method for estimating small area parameters under an inverse Gaussian model. The method is extended to estimate small area parameters for finite populations. The Gibbs sampler is proposed as a mechanism for implementing the Bayesian paradigm. We illustrate the method by application to household income survey data, comparing it against the usual lognormal model for positive...

This paper presents our first attempts to develop a new methodology for measuring housing deficit at small areas. It combines the advantages of two types of census data: (a) individual-level sample data, which are very useful for depicting many dimensions of the housing deficit, but do not present detailed geographic information; and (b) universal data with detailed spatial resolution (census t...

Introduction National statistical bureaus often provide estimates of different small area indicators (e.g., unemployment, average income) at different geographical levels which have been computed using different methods. Spatio-temporal models, for example, take into account different geographic and temporal structures of the data in order to improve estimation. The purpose is to borrow strengt...

Reducing child undernutrition is a key social policy objective of the Ethiopian government. Despite substantial reduction over the last decade and a half, child undernutrition is still high; with 48 percent of children either stunted, underweight or wasted, undernutrition remains an important child health challenge. The existing literature highlights that targeting of efforts to reduce undernut...

In this paper a regression model is developed for the post-censal estimation of the population sizes of small-areas. The approach is nonstochastic. It is assumed that current population sizes have been determined by a function of those obtained at the last census, together with the associated values of certain symptomatic variables. As well, the current values of these variables are assumed to ...

The poverty mapping methodology for estimating welfare rankings from small areas has proven to be useful in guiding allocation of government funds, regional planning, and general policy formulation. Nevertheless, poverty mapping also suffers from a series of by now well recognized shortcomings. We apply an approach based on first order dominance (FOD) to small area estimation. Five advantages t...

We propose a new small area estimation approach that combines small area random effects with a smooth, nonparametrically specified trend. By using penalized splines as the representation for the nonparametric trend, it is possible to express the small area estimation problem as a mixed effect regression model. We show how this model can be fitted using existing model fitting approaches such as ...

Consider the small area estimation when positive area-level data like income, revenue, harvests or production are available. Although a conventional method is the logtransformed Fay-Herriot model, the log-transformation is not necessarily appropriate. Another popular method is the Box-Cox transformation, but it has drawbacks that the maximum likelihood estimator (ML) of the transformation param...

Small area estimators commonly borrow strength from other related areas. These indirect estimators use models (explicit or implicit) that relate the small areas through supplementary data. Various unit-level and area-level small area models are proposed in the literature, but all these models assume the small area mean is linearly related with supplementary information. In this article, we prop...

Title of dissertation: THE BAYESIAN AND APPROXIMATE BAYESIAN METHODS IN SMALL AREA ESTIMATION Santanu Pramanik, Doctor of Philosophy, 2008 Dissertation directed by: Professor Partha Lahiri Joint Program in Survey Methodology For small area estimation, model based methods are preferred to the traditional design based methods because of their ability to borrow strength from related sources. The i...