Model robustness for simultaneous confidence bands

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Confidence bands for the simple linear model are examined to assess their degrees of robustness to departures from the model. All calculations are made under an interval constraint on the range of interest for the predictor variable. The true model is taken to be a quadratic polynomial, and departure from the linear case is considered in terms of increasing magnitude of a curvature parameter. Proposed measures of robustness include the actual coverage probability under the true model and a measure of percentage coverage over the predictor variable axis when the band fails to cover the true quadratic model. The different band functions considered include hyperbolic bands constructed from Scheffd’s 5 method, linear-segmented bands, and fixed-width (minimax) and minimax-regret bands. In terms of preserving coverage probability under quadratic mis-specification, the fixed-width and linear-segment bands perform best, the former being preferred when the constraint interval on the predictor variable is small. When coverage is lost over some portion of the constraint interval, the fixed-width bands are also shown to preserve the greatest percentage of covered (predictor) axis values. The favorable performance of the fixed-width bands may be due to their generally wider, more rigid shape, relative to more curvilinear competitors. This may allow the bands to retain more extreme quadratic misspecifications than other bands.

Original languageEnglish (US)
Pages (from-to)879-885
Number of pages7
JournalJournal. American Statistical Association
Volume82
Issue number399
DOIs
StatePublished - 1987
Externally publishedYes

Fingerprint

Model Robustness
Simultaneous Confidence Bands
Predictors
Misspecification
Coverage Probability
Minimax
Interval
Percentage
Coverage
Bandwidth
Robustness
Confidence Bands
Regret
Confidence
Quadratic Polynomial
Model
Linear Model
Extremes
Curvature
Cover

Keywords

  • Coverage probability
  • Mean axis coverage
  • Quadratic regression
  • Simple linear regression

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Model robustness for simultaneous confidence bands. / Piegorsch, Walter W.

In: Journal. American Statistical Association, Vol. 82, No. 399, 1987, p. 879-885.

Research output: Contribution to journalArticle

@article{322eee7ab386470f9b78f42d70b4b647,
title = "Model robustness for simultaneous confidence bands",
abstract = "Confidence bands for the simple linear model are examined to assess their degrees of robustness to departures from the model. All calculations are made under an interval constraint on the range of interest for the predictor variable. The true model is taken to be a quadratic polynomial, and departure from the linear case is considered in terms of increasing magnitude of a curvature parameter. Proposed measures of robustness include the actual coverage probability under the true model and a measure of percentage coverage over the predictor variable axis when the band fails to cover the true quadratic model. The different band functions considered include hyperbolic bands constructed from Scheffd’s 5 method, linear-segmented bands, and fixed-width (minimax) and minimax-regret bands. In terms of preserving coverage probability under quadratic mis-specification, the fixed-width and linear-segment bands perform best, the former being preferred when the constraint interval on the predictor variable is small. When coverage is lost over some portion of the constraint interval, the fixed-width bands are also shown to preserve the greatest percentage of covered (predictor) axis values. The favorable performance of the fixed-width bands may be due to their generally wider, more rigid shape, relative to more curvilinear competitors. This may allow the bands to retain more extreme quadratic misspecifications than other bands.",
keywords = "Coverage probability, Mean axis coverage, Quadratic regression, Simple linear regression",
author = "Piegorsch, {Walter W}",
year = "1987",
doi = "10.1080/01621459.1987.10478512",
language = "English (US)",
volume = "82",
pages = "879--885",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "399",

}

TY - JOUR

T1 - Model robustness for simultaneous confidence bands

AU - Piegorsch, Walter W

PY - 1987

Y1 - 1987

N2 - Confidence bands for the simple linear model are examined to assess their degrees of robustness to departures from the model. All calculations are made under an interval constraint on the range of interest for the predictor variable. The true model is taken to be a quadratic polynomial, and departure from the linear case is considered in terms of increasing magnitude of a curvature parameter. Proposed measures of robustness include the actual coverage probability under the true model and a measure of percentage coverage over the predictor variable axis when the band fails to cover the true quadratic model. The different band functions considered include hyperbolic bands constructed from Scheffd’s 5 method, linear-segmented bands, and fixed-width (minimax) and minimax-regret bands. In terms of preserving coverage probability under quadratic mis-specification, the fixed-width and linear-segment bands perform best, the former being preferred when the constraint interval on the predictor variable is small. When coverage is lost over some portion of the constraint interval, the fixed-width bands are also shown to preserve the greatest percentage of covered (predictor) axis values. The favorable performance of the fixed-width bands may be due to their generally wider, more rigid shape, relative to more curvilinear competitors. This may allow the bands to retain more extreme quadratic misspecifications than other bands.

AB - Confidence bands for the simple linear model are examined to assess their degrees of robustness to departures from the model. All calculations are made under an interval constraint on the range of interest for the predictor variable. The true model is taken to be a quadratic polynomial, and departure from the linear case is considered in terms of increasing magnitude of a curvature parameter. Proposed measures of robustness include the actual coverage probability under the true model and a measure of percentage coverage over the predictor variable axis when the band fails to cover the true quadratic model. The different band functions considered include hyperbolic bands constructed from Scheffd’s 5 method, linear-segmented bands, and fixed-width (minimax) and minimax-regret bands. In terms of preserving coverage probability under quadratic mis-specification, the fixed-width and linear-segment bands perform best, the former being preferred when the constraint interval on the predictor variable is small. When coverage is lost over some portion of the constraint interval, the fixed-width bands are also shown to preserve the greatest percentage of covered (predictor) axis values. The favorable performance of the fixed-width bands may be due to their generally wider, more rigid shape, relative to more curvilinear competitors. This may allow the bands to retain more extreme quadratic misspecifications than other bands.

KW - Coverage probability

KW - Mean axis coverage

KW - Quadratic regression

KW - Simple linear regression

UR - http://www.scopus.com/inward/record.url?scp=30944439660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30944439660&partnerID=8YFLogxK

U2 - 10.1080/01621459.1987.10478512

DO - 10.1080/01621459.1987.10478512

M3 - Article

VL - 82

SP - 879

EP - 885

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 399

ER -