Modeling malaria incidence in Bengkulu province using small area estimation

The purpose of this study is to model malaria incidence in Bengkulu province using small area estimation. The model used is the Log-Normal model. In this research, the method is applied to estimate parameters in Small Area Estimation using Hierarchical Bayesian and direct estimation methods. The data used was collected by the Bureau of Statistics (BPS). The results show that the parameters for the discrete data in the Small Area Estimation, which is the average estimation of the Log-Normal prior function, are more accurate than the direct estimation. Other results are obtained from these estimates of the Hierarchical Bayesian estimator have a trend (tendency) which is equal to the direct estimator. It means that both methods generate consistent estimators.