We designed this package to provide several functions for area level of small area estimation using hierarchical Bayesian (HB) method. This package provides model using panel data for variable interest.This package also provides a dataset produced by a data generation. The 'rjags' package is employed to obtain parameter estimates. Model-based estimators involves the HB estimators which include the mean and the variation of mean. For the reference, see Rao and Molina (2015).
We designed this package to provide several functions for area level of small area estimation using hierarchical Bayesian (HB) method. This package provides model using Rao-Yu Model for variable interest.This package also provides a dataset produced by a data generation. The “rjags” package is employed to obtain parameter estimates. Model-based estimators involves the HB estimators which include the mean and the variation of mean. For the reference, see Rao and Molina (2015) and Torabi and Shokoohi (2012).
Velia Tri Marliana, Azka Ubaidillah
Velia Tri Marliana [email protected]
RaoYuAr1() This function gives estimation of y using Hierarchical
Bayesian under Rao Yu ModelPanel() This function gives estimation of y using Hierarchical
Bayesian under Rao Yu Model when rho = 0You can install the development version of saeHB.panel from GitHub with:
# install.packages("devtools")
devtools::install_github("Veliatrimarliana/saeHB.panel")
This is a basic example which shows you how to solve a common problem:
library(saeHB.panel)
data(dataAr1)
formula = ydi ~ xdi1 + xdi2
area = max(dataAr1[, "area"])
period = max(dataAr1[,"period"])
vardir = dataAr1[,4]
result <- Raoyu.Ar1(formula, area, period, vardir = vardir, data = dataAr1)
Extract area mean estimation
result$Est
Extract coefficient estimation
result$coefficient
Extract area random effect variance
result$refVar
##References * Rao, J.N.K & Molina. (2015). Small Area Estimation 2nd Edition. New York: John Wiley and Sons, Inc. * Torabi, M., & Shokoohi, F. (2012). Likelihood inference in small area estimation by combining time-series and cross-sectional data. Journal of Multivariate Analysis, 111, 213–221. https://doi.org/10.1016/j.jmva.2012.05.016