Nonparametric relative error regression estimator under η-weak dependence with geometric rate

Abstract

In this paper, we present a non-parametric regression approach to study the relationship between two variables. The relative error estimator serves as a more robust alternative to classical regression, particularly under the influence of outliers. First, we present the relative error regression estimator. Next, we then investigate both pointwise and uniform almost complete convergence, along with their rates, under the η-weak dependence condition. Finally, the numerical simulations demonstrate the effectiveness and accuracy of the proposed estimator.