
A Bayesian Reweighting Technique for Small Area Estimation 513
in CO [35]. It can adopt the g e neralised model operated in the GREGWT to
link the sample and unobserved units in the population. In contrast, from the
viewpoint of CO, this method us e s the MCMC s imulation with a posterior
density based iterative algorithm. As the joint posterior probabilities of pa-
rameters for the sample units and unobserved population units are estimated
through MCMC, this microda ta simulation methodology is somewhat linked
with a chain Monte Carlo sampling. However, it is rather different from the
multiple imputation technique advanced by Rubin [30] and o thers. ...