1983•NORA - Norwegian Open Research ArchivesOpen access

UNIFORM CONFIDENCE BOUNDS FOR NONPARAMETRIC REGRESSION

Steinar Bjerve, Kjell A. Doksum, Brian S. Yandell

Open full text 0 citations

Abstract

Let (X, Y) be a bivariate random variable and let m(x) = E(YIX = x) be the regression function of Y on X. Suppose that v 1 , .••,Yn are independent observations of Y at X= x 1 , .••,xn.We consider nearest neighbor estimates, m(x), and employ well-known inequalities to obtain exact and asymptotic uniform confidence bounds for Em(x) and m(x) based on m(x).Finally we discuss bias-properties of m(x).

Open-access reader

About this research paper

What this paper is about

Let (X, Y) be a bivariate random variable and let m(x) = E(YIX = x) be the regression function of Y on X. Suppose that v 1 , .••,Yn are independent observations of Y at X= x 1 , .••,xn.We consider nearest neighbor estimates, m(x), and employ well-known inequalities to obtain exact and asymptotic uniform confidence bounds for Em(x) and m(x) based on m(x).Finally we discuss bias-properties of m(x).

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Let (X, Y) be a bivariate random variable and let m(x) = E(YIX = x) be the regression function of Y on X. Suppose that v 1 , .••,Yn are independent observations of Y at X= x 1 , .••,xn.We consider nearest neighbor estimates, m(x), and employ well-known inequalities to obtain exact and asymptotic uniform confidence bounds for Em(x) and m(x) based on m(x).Finally we discuss bias-properties of m(x).

Key concepts: Nonparametric statistics, Statistics, Regression, Nonparametric regression, Confidence interval, Regression analysis, Econometrics, Mathematics

Related papers

Back to paper searchBrowse research topicsOriginal source
UNIFORM CONFIDENCE BOUNDS FOR NONPARAMETRIC REGRESSION — Research Paper | ScholarLens