A global smoothing procedure is developed using B-spline function approximations for estimating the unknown functions of a varying coefficient model with repeated measurements. general formulation used to treat mean, median, quantile and robust mean regressions in one setting. The convergence rates M-estimators are established. asymptotic distributes derived approximate confidence intervals also Various applications main results, including conditional robustifying regression given. Finite...

This paper considers an estimation of semiparametric functional (varying)-coefficient quantile regression with spatial data. A general robust framework is developed that treats for data in a natural way. The local M-estimators the unknown functional-coefficient functions are proposed by using linear approximation, and their asymptotic distributions then established under weak mixing conditions allowing processes to be either stationary or nonstationary trends. Application soil set...

This paper studies estimation in semi-functional linear regression. A general formulation is used to treat mean regression, median quantile regression and robust one setting. The slope function estimated by the functional principal component basis nonparametric approximated a B-spline function. global convergence rates of estimators unknown are established under suitable norm. rate mean-squared prediction error for proposed also established. Finite sample properties our procedures studied...

Abstract The varying coefficient model is a useful extension of linear models and has many advantages in practical use. To estimate the unknown functions model, kernel type with local least-squares (L 2) estimation methods been proposed by several authors. When data contain outliers or come from population heavy-tailed distributions, L 1-estimation should yield better estimators. In this article, we present method derive asymptotic distributions 1-estimators. simulation results for two...

This paper considers a varying-coefficient partially linear regression with spatial data. A global smoothing procedure is developed by using B-spline function approximations for estimating the unknown parameters and coefficient functions. Under mild regularity assumptions, asymptotic distribution of estimator parameter vector established. The convergence rates estimators functions are distributions also derived. Finite sample properties our procedures studied through Monte Carlo simulations....

This paper proposes nonparametric estimation methods for functional linear semiparametric quantile regression, where the conditional of scalar responses is modelled by both and covariates an additional unknown function term. The slope estimated using principal component basis approximated a piecewise polynomial function. asymptotic distribution estimators parameters derived global convergence rate estimator established under suitable norm. also established. Simulation studies are conducted...

A global smoothing procedure is developed using B-spline function approximation for estimating the unknown functions of a functional coefficient regression model with spatial data. general formulation used to treat mean regression, median quantile and robust in one setting. The convergence rates estimators are established. Various applications main results, including conditional robustifying given. Finite sample properties our procedures studied through Monte Carlo simulations. housing data...

This paper studies M-estimation in functional linear regression which the dependent variable is scalar while covariate a function. An estimator for slope function obtained based on principal component basis. The global convergence rate of M-estimator unknown established. mean-squared prediction error proposed estimators also Monte Carlo simulation are conducted to examine finite-sample performance procedure. Finally, method applied analyze Berkeley growth data.

This article considers a nonparametric varying coefficient regression model with longitudinal observations. The relationship between the dependent variable and covariates is assumed to be linear at specific time point, but coefficients are allowed change over time. A general formulation used treat mean regression, median quantile robust in one setting. local M-estimators of unknown functions obtained by method. asymptotic distributions both interior boundary points established. Various...

Data collected on the surface of earth often have spatial interaction. In this paper, a global smoothing procedure is developed using tensor product B-spline function approximations for estimating multi-dimensional conditional regression function. Under mild regularity assumptions, convergence rates estimators are established. Asymptotic results show that our achieve optimal rate. The asymptotic normality estimator also derived. Finite sample properties procedures studied through Monte Carlo...

文献 [13] 研究了纵向数据变系数部分线性模型的分位数估计. 同均值回归相比, 分位数回归具有以下 几方面的优势: 第一