Spatial Empirical Best Linear Unbiased Prediction Methods for Small Areas to Estimate monthly Expenditure per capita (Case Study: The Province of East Java District Jember).

Abstract

Poverty of an area can be measured by their per capita expenditure. Direct estimation of expenditure per capita based on small area data may cause a large variance of estimates. Spatial Empirical Best Linear Unbiased Prediction (SEBLUP) is one of the small area estimation methods to solve the problems. The method is developed from EBLUP allowing correlation among area. Susenas Data was used to compare SEBLUP, direct estimation and EBLUP. The result showed that the standard error of SEBLUP was smaller than those of direct estimation and EBLUP