학술논문
Varying coefficient single-index regression model with missing responses under rank-based modelling.
Document Type
Journal
Author
Otlaadisa, Masego (BT-BIUST2-MS) AMS Author Profile; Bindele, Huybrechts F. (1-SAL) AMS Author Profile; Abebe, Asheber (1-ABRN-MST) AMS Author Profile; Correia, Hannah (1-HRV-DTI) AMS Author Profile
Source
Subject
62 Statistics -- 62F Parametric inference
62F12Asymptotic properties of estimators
62Statistics -- 62G Nonparametric inference
62G05Estimation
62G20Asymptotic properties
62F12
62
62G05
62G20
Language
English
Abstract
In this paper, the authors consider a varying coefficients single-index regression model with responses missing at random. They propose rank-based estimators of the index coefficient and the functional coefficients and study their asymptotic properties, including consistency and asymptotic normality under mild conditions. To demonstrate the performance of the proposed approach, Monte Carlo simulation experiments are carried out that show that the proposed approach provides robust and more efficient estimators compared to its least-squares counterpart. Also, classic methods of proving central limit theorems are skillfully used to give rigorous proof.