Small area estimation using skew normal models

Ferraz, V. R. S.; Moura, Fernando A. S.

doi:10.1016/j.csda.2011.07.005

Cited by 19 publications

(22 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4/ 3.1. The joint skew normal and skew t normal models Ferraz and Moura (2012) proposed the following joint model for the direct survey estimator y i and its sampling variance estimatorφ i :…”

Section: The Skew Area Level Modelsmentioning

confidence: 99%

“…Another alternative is to use a non-symmetric distribution to model the direct estimator. Along this line, we refer to the work of Ferraz and Moura (2012), who modelled the direct survey estimator as skew normally distributed to take into account the skewed data. Nevertheless, in their formulation, as the area sample size increases, it converges to the commonly employed normal model.…”

Section: Introductionmentioning

confidence: 99%

“…In fact, owing to the central limit theorem, as the sample size increases, any weighted average estimator's distribution must uniformly converge to a normal distribution. Ferraz and Moura (2012) successfully fitted the skew normal model to head-of-household mean income for 140 enumeration areas. The data were extracted from an experimental Brazilian demographic census.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Small Area Models for Skewed Brazilian Business Survey Data

Moura

Neves

Silva

2017

Journal of the Royal Statistical Society Series A: Statistics in Society

View full text Add to dashboard Cite

Summary The Brazilian Institute of Geography and Statistics performs an annual service survey that focuses on segments of the tertiary sector. Sample estimates for some economic activities in the north, north‐east and midwest regions of Brazil have low precision due to the sample design. Furthermore, one of the main variables of interest is considerably skewed with potential outliers. To overcome this problem, skew normal and skew t‐models are proposed to produce model‐based estimates. The small domain estimation models relate operating revenue variables to potential auxiliary variables (the number of employed people and wages) obtained from a business register. The models proposed are compared with the usual Fay–Herriot model under the assumptions of known and unknown sampling variances and its transformed log‐version under the assumption of known variances. The evaluation studies with real business survey data show that the models proposed seem to be more efficient for small area predictions under skewed data than the customarily employed Fay–Herriot model, as well as its log‐normal version.

show abstract

Section: The Skew Area Level Modelsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Small Area Models for Skewed Brazilian Business Survey Data

Moura

Neves

Silva

2017

Journal of the Royal Statistical Society Series A: Statistics in Society

View full text Add to dashboard Cite

show abstract

“…(a) hierarchically generating samples of the skew normal density by using the stochastic representation (Henze, 1986;Ferraz and Moura, 2011); (b) explicitly writing the skew normal density formula into the BUGS code, which can be done by using what is known as 'the trick for specifying new distributions' (Spiegelhalter et al, 2002).…”

Section: Prediction Strategymentioning

confidence: 99%

Small Domain Estimation of Business Statistics by Using Multivariate Skew Normal Models

Ferrante

Pacei

2017

Journal of the Royal Statistical Society Series A: Statistics in Society

View full text Add to dashboard Cite

Summary Small domain business statistics are becoming important for better planning business policies. We focus on the estimation of the averages of value added and labour cost in small domains. To take into account the positive skewness in the distribution of outcomes and the correlation between them, we propose a bivariate skew normal small area model. Estimates are obtained from real survey data. The performance of the estimator proposed is evaluated on the basis of both survey data and a synthetic firm population. Results show that the model proposed increases the estimates’ reliability and that the estimates obtained make it possible to perform detailed regional economic studies.

show abstract

“…Frühwirth-Schnatter and Pyne (2010) have recently proposed a fully Bayesian analysis of a mixture of skew-normal and skew-t densities. Other recent and important advances in the application of multivariate skew-normal models can be found in Fung and Seneta (2010), Panagiotelis and Smith (2010), Ferraz and Moura (2012) and Cabral et al (2012). The computational approach described in Frühwirth-Schnatter and Pyne (2010) differs from ours in two respects: i) they adopt conjugate priors in order to facilitate a Gibbs sampling strategy for simulating from the posterior; ii) as a consequence of i), we adopt a different sampling strategy, based on importance sampling rather than MCMC; we will describe the PMC algorithm in detail in Section 4.…”

Section: Motivationsmentioning

confidence: 99%

Bayesian inference for the multivariate skew-normal model: A population Monte Carlo approach

Liseo

Parisi

2013

Computational Statistics & Data Analysis

View full text Add to dashboard Cite

Frequentist and likelihood methods of inference based on the multivariate skewnormal model encounter several technical difficulties with this model. In spite of the popularity of this class of densities, there are no broadly satisfactory solutions for estimation and testing problems. A general population Monte Carlo algorithm is proposed which: 1) exploits the latent structure stochastic representation of skew-normal random variables to provide a full Bayesian analysis of the model and 2) accounts for the presence of constraints in the parameter space. The proposed approach can be defined as weakly informative, since the prior distribution approximates the actual reference prior for the shape parameter vector. Results are compared with the existing classical solutions and the practical implementation of the algorithm is illustrated via a simulation study and a real data example. A generalization to the matrix variate regression model with skew-normal error is also presented.

show abstract

Small area estimation using skew normal models

Cited by 19 publications

References 9 publications

Small Area Models for Skewed Brazilian Business Survey Data

Small Area Models for Skewed Brazilian Business Survey Data

Small Domain Estimation of Business Statistics by Using Multivariate Skew Normal Models

Bayesian inference for the multivariate skew-normal model: A population Monte Carlo approach

Contact Info

Product

Resources

About