Berry-Esseen bounds for M-estimators in censored regression models under (Λ ₁ (E), ψ, ϵ) -weak dependence

This paper addresses the importance of analyzing left truncated and right censored (LTRC) data, which plays a critical role in fields such as survival analysis and econometrics. We establish Berry–Esseen bounds for an M-estimator of the regression function under the assumption of stationary (Λ1(E),ψ,ϵ)-weak dependence random variables. A key innovation of our work lies in the use of Bernstein blocks methods to derive these bounds, representing a novel approach in the study of estimators within the LTRC strong mixing framework. This method offers new insights and contributes to the advancement of research in this area. To support the theoretical findings, the proposed estimators are applied to the Stanford2 dataset to demonstrate their finite sample performance.