Regression with imprecise data: A robust approach, in

We introduce a robust regression method for impre-cise data, and apply it to social survey data. Our method combines nonparametric likelihood inference with imprecise probability, so that only very weak as-sumptions are needed and different kinds of uncer-tainty can be taken into account. The proposed re-gression method is based on interval dominance: in-terval estimates of quantiles of the error distribution are used to identify plausible descriptions of the rela-tionship of interest. In the application to social sur-vey data, the resulting set of plausible descriptions is relatively large, reflecting the amount of uncertainty inherent in the analyzed data set