Quantitative bounds for concentration-of-measure inequalities and empirical regression: the independent case (Q2422732)

This paper is concerned with the study of the deviation of the average \(L^2\)-error associated to the least squares regressor over a family of functions, obtained from independent but not necessarily identically distributed samples of explanatory and response variables, from the minimal average \(L^2\)-error associated with this family of functions. The authors obtain nonasymptotic deviation inequalities that generalize and refine corresponding results in the independent identically distributed case.