Adaptive nonparametric instrumental regression by model selection. arxiv:1003.3128v1, Université Catholique de Louvain

We consider the problem of estimating the structural function in nonpara-metric instrumental regression, where in the presence of an instrument W a response Y is modeled in dependence of an endogenous explanatory variable Z. The proposed estimator is based on dimension reduction and additional thresh-olding. The minimax optimal rate of convergence of the estimator is derived assuming that the structural function belongs to some ellipsoids which are in a certain sense linked to the conditional expectation operator of Z given W. We il-lustrate these results by considering classical smoothness assumptions. However, the proposed estimator requires an optimal choice of a dimension parameter de-pending on certain characteristics of the unknown structural function and the conditional expectation operator of Z given W, which are not known in practice. The main issue addressed in our work is a fully adaptive choice of this dimension parameter using a model selection approach under the restriction that the condi-tional expectation operator of Z given W is smoothing in a certain sense. In this situation we develop a penalized minimum contrast estimator with randomized penalty and collection of models. We show that this data-driven estimator can attain the lower risk bound up to a constant over a wide range of smoothness classes for the structural function