MathLAB Journal

MISSING AT RANDOM IN NONPARAMETRIC REGRESSION FOR FUNCTIONAL STATIONARY ERGODIC DATA IN THE FUNCTIONAL INDEX MODEL

Abstract

FATIMA AKKAL, MUSTAPHA MEGHNAFI AND ABBES RABHI

The main objective of this paper is to estimate non-parametrically the the estimator for the regression function operator when the observations are linked with a single-index. The functional stationary ergodic data with missing at random (MAR) are considered. In particular, we construct the kernel type estimator of the regression operator, some asymptotic properties such as the convergence rate in probability as well as the asymptotic normality of the estimator are established under some mild conditions respectively. As an application, the asymptotic (1 − ) confidence interval of the regression operator is also presented for 0 <  < 1.<