Estimation of Poverty in Small Areas

Abstract

A qualitative techniques of poverty estimation is needed to better implement, monitor and determine national areas where support is most required. The problem of small area estimation (SAE) is the production of reliable estimates in areas with small samples. The precision of estimates in strata deteriorates (i.e. the precision decreases when the standard deviation increases), if the sample size is smaller. In these cases traditional direct estimators may be not precise and therefore pointless. Currently there are many indirect methods for SAE. The purpose of this paper is to analyze several diff erent types of techniques which produce small area estimates of poverty.

Keywords

• Poverty
• small area estimation
• Horvitz-Thompson
• Generalised Regression
• Synthetic
• Jack-Knife