Non-Parametric Regression and Riesz Estimators

In this paper, we consider a non-parametric regression model relying on Riesz estimators. This linear is similar to the usual since they both rely projection operators. We indicate that estimator relies positive basis elements of finite-dimensional sub-lattice generated by rows some design matrix. A strong motivation for using data explanatory variables may come from categorical variables. Calculations related are very easy arise measurability in probability spaces. Moreover, show fitted estimators an ordinary least squares model. Any vector Euclidean space supposed be rendom variable under objective values, being used expected utility theory and its applications. Finally, reader notice goodness-of-fit measures those defined regression. Due fact non-parametric, it include samples relevant finance actuarial science