Hotspot Detection and Mapping of Poverty

In this paper we aim at detecting poverty hotspots and preparing the corresponding maps. Poverty and inequality maps — spatial descriptions of the distribution of poverty — are most useful to policymakers and researchers when they are finely disaggregated, i.e. when they represent small geographic units, such as cities, municipalities, regions or other administrative partitions of a country. Moreover, when poverty hotspots are detected, policymakers can use them to propose appropriate programmes and anti-poverty policies. We demonstrate the construction of detailed poverty maps, primarily using data from a Population Census in conjunction with an intensive small-scale national sample survey. The methodology adopted combines census and survey information to produce finely disaggregated maps. The basic idea is to estimate a linear regression model with local variance components using information from the smaller and richer sample data in the case of Albania the Living Standard Measurement Study (LSMS) conducted in 2002 — in conjunction with the large-scale but limited information from the 2001 Population and Housing Census.