Evaluating different approaches to estimating vulnerability

摘要: A number of researchers have recently proposed a variety different `vulnerability' measures designed to capture the welfare consequences risk for poor households, and also approaches estimating these various household vulnerability. However, it's possible `mix-and-match' estimators measures. Here we conduct Monte Carlo experiments explore performance with measures, under assumptions regarding underlying economic environment. We find that when environment is stationary, consumption expenditures are measured without error, best estimator one by Chaudhuri (2001), regardless what measure vulnerability employed. If risk-sensitive, but simple authors (2003) generally performs best. distribution non-stationary, modification an Pritchett et al. (2000) Future research should focus on combining efficiency Chauduri good properties (in environments measurement error), non-stationary environments) estimators. even present technology simple, much more informative, useful than static poverty provided has at least two rounds panel data.