Title: Quantile regression with an endogenous misclassified binary regressor
Authors: Carlos Lamarche - University of Kentucky (United States) [presenting]
Abstract: Misreporting of participation in social programs is common, and it has been increasing in all major surveys. We investigate the estimation of a quantile regression model with endogenous misreporting. We propose a two-step approach and show that the estimator is consistent and asymptotically normal. The identification of the model relies on a parametric first stage and the use of additional measurements including instrumental variables. Simulation studies offer small sample behavior of the proposed estimator in comparison with other approaches. An illustration of the new approach using survey data is considered.