Bias correction for underreported data in small area mapping.

Data quality is emerging as an essential characteristics of all data driven processes. The problem is particularly severe when health or vital statistics are concerned, with important consequences on government intervention policies and distribution of financial resources. In this paper, we deal with the underreporting issue with particular attention on its effects on the estimation of the prevalence of a phenomenon. We propose a non parametric compound Poisson model that allows for the estimation of underreporting probabilities. We will apply the proposed model to original data about the incidence of Chronic Kidney Disease (CKD) in Apulia.