PovMap is based on a small area estimation procedure developed by the World Bank (Lanjouw and Elber, 2002, 2003). This method uses survey data to establish an income/expenditure prediction model, t. Then applies it to a compatible census data set at the household level. The predicted income/expenditure is further aggregated to compute poverty/inequality measurements (FTG0, FGT, FGT2, Gini, etc.) at the community level (such as township, county or district). In some specific situations, non-census data can also be used. The intermediate estimation of household income/expenditure of each simulation can also been kept.
PovMap 2.0 provides computational solutions to all stages of poverty mapping activities. It uses a proprietary data engine to ensure the speedy processing of census data. Its functions include data processing, distributional comparison, regression analysis, univariate analysis, correlation analysis, cross tabulation and simulation. PovMap 2.1 is an experimental package for nutrition mapping. PovMap 1.1, or earlier, are for the simulation stage of poverty mapping.
Peter Lanjouw on Mapping Poverty (Video, June 25, 2012, 2 minutes)