Coverage Accuracy of Confidence Intervals in Nonparametric Regression

Coverage Accuracy of Confidence Intervals in Nonparametric Regression Point-wise confidence intervals for a nonparametric regression function with random design points are considered. The confidence intervals are those based on the traditional normal approximation and the empirical likelihood. Their coverage accuracy is assessed by developing the Edgeworth expansions for the coverage probabilities. It is shown that the empirical likelihood confidence intervals are Bartlett correctable. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Mathematicae Applicatae Sinica Springer Journals

Coverage Accuracy of Confidence Intervals in Nonparametric Regression

Coverage Accuracy of Confidence Intervals in Nonparametric Regression

Acta Mathematicae Applicatae Sinica, English Series Vol. 19, No. 3 (2003) 387–396 Coverage Accuracy of Confidence Intervals in Nonparametric Regression 1 2 Song-xi Chen , Yong-song Qin Department of Statistics and Applied Probability, National University of Singapore, Singapore 117543, Singapore (Email: stacsx@nus.edu.sg) Department of Mathematics, Guangxi Normal University, Guilin 541004, China (Email: ysqin@eyou.com) Abstract Point-wise confidence intervals for a nonparametric regression function with random design points are considered. The confidence intervals are those based on the traditional normal approximation and the empirical likelihood. Their coverage accuracy is assessed by developing the Edgeworth expansions for the coverage probabilities. It is shown that the empirical likelihood confidence intervals are Bartlett correctable. Keywords Confidence interval, empirical likelihood, Nadaraya-Watson estimator, normal approximation 2000 MR Subject Classification 62G05, 62E20 1 Introduction d+1 d Let (X ,Y ),··· , (X ,Y ) be independent copies of R random vectors (X, Y )where X ∈ R 1 1 n n with density f.Let m(x)= E(Y |X = x) be the conditional mean function and σ (x)= Var(Y |X = x) be the conditional variance function of Y given X = x.For any x, u ∈ R ,let 2 2 2 V (u|x)= E[{Y − m(x)} |X = u]. Clearly, V (u|x)= σ (u)+ {m(u) − m(x)} . The Nadaraya-Watson estimator for m(x)at any given x is K (x − X )Y h i i m  (x)= , (1.1) K (x − X ) h i where K (t)= K(t/h), K is a kernel function and h is the smoothing bandwidth. This paper is concerned with the construction of point-wise confidence intervals for m(x) at any fixed x in conjunction with the Nadaraya-Watson estimator when the design points are random. Confidence intervals based on the asymptotic normality of the Nadaraya-Watson [8] estimator and the percentile bootstrap are reviewed in [11]. Hall considers coverage accuracy of...
Loading next page...
 
/lp/springer_journal/coverage-accuracy-of-confidence-intervals-in-nonparametric-regression-3z0HzIL15s
Publisher
Springer Berlin Heidelberg
Copyright
Copyright © 2003 by Springer-Verlag Berlin Heidelberg
Subject
Mathematics; Applications of Mathematics; Math Applications in Computer Science; Theoretical, Mathematical and Computational Physics
ISSN
0168-9673
eISSN
1618-3932
D.O.I.
10.1007/s10255-003-0113-3
Publisher site
See Article on Publisher Site

Abstract

Point-wise confidence intervals for a nonparametric regression function with random design points are considered. The confidence intervals are those based on the traditional normal approximation and the empirical likelihood. Their coverage accuracy is assessed by developing the Edgeworth expansions for the coverage probabilities. It is shown that the empirical likelihood confidence intervals are Bartlett correctable.

Journal

Acta Mathematicae Applicatae SinicaSpringer Journals

Published: Mar 3, 2017

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from Google Scholar, PubMed
Create lists to organize your research
Export lists, citations
Access to DeepDyve database
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off