Inverse Gaussian Model for Small Area Estimation via Gibbs Sampling
نویسندگان
چکیده
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 positively skewed data. Key words/phrases: Finite population sampling, hierarchical Bayesian inference, lognormal model, MCMC integration, shrinkage estimates
منابع مشابه
Robust Factor Analysis Using the Multivariate t-Distribution
Factor analysis is a standard method for multivariate analysis. The sampling model in the most popular factor analysis is Gaussian and has thus often been criticized for its lack of robustness. A simple robust extension of the Gaussian factor analysis model is obtained by replacing the multivariate Gaussian distribution with a multivariate t-distribution. We develop computational methods for bo...
متن کاملThe effect of estimation methods on fractal modeling for anomalies’ detection in the Irankuh area, Central Iran
This study aims to recognize effect of Ordinary Kriging (OK) and Inverse Distance Weighted (IDW) estimation methods for separation of geochemical anomalies based on soil samples using Concentration-Area (C-A) fractal model in Irankuh area, central Iran. Variograms and anisotropic ellipsoid were generated for the Pb and Zn distribution. Thresholds values from the C-A log-log plots based on the e...
متن کاملThe Bayesian Lasso
The Lasso estimate for linear regression parameters can be interpreted as a Bayesian posterior mode estimate when the priors on the regression parameters are independent double-exponential (Laplace) distributions. This posterior can also be accessed through a Gibbs sampler using conjugate normal priors for the regression parameters, with independent exponential hyperpriors on their variances. T...
متن کاملParameter estimation for non-Gaussian autoregressive processes
It is proposed to jointly estimate the parameters of nonGaussian autoregressive (AR) processes in a Bayesian context using the Gibbs sampler. Using the Markov chains produced by the sampler an approximation to the vector MAP estimator is implemented. The results reported here used AR(4) models driven by noise sequences where each sample is iid as a two component Gaussian sum mixture. The result...
متن کاملComparison of Maximum Likelihood Estimation and Bayesian with Generalized Gibbs Sampling for Ordinal Regression Analysis of Ovarian Hyperstimulation Syndrome
Background and Objectives: Analysis of ordinal data outcomes could lead to bias estimates and large variance in sparse one. The objective of this study is to compare parameter estimates of an ordinal regression model under maximum likelihood and Bayesian framework with generalized Gibbs sampling. The models were used to analyze ovarian hyperstimulation syndrome data. Methods: This study use...
متن کامل