Constant width simultaneous confidence bands in multiple linear regression with predictor variables constrained in intervals

W. Liu, M. Jamshidian, Ying Zhang, F. Bretz

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

In the last fifty years, a great deal of research effort has been made on the construction of simultaneous confidence bands for a linear regression function. Two most frequently quoted confidence bands in the statistics literature are the Scheffé type and constant width bands over a given rectangular region of the predictor variables. For the constant width bands, a method is given by Gafarian [Gafarian, A.V., 1964, Confidence bands in straight line regression. Journal of the American Statistical Association, 59, 182-213.] for the calculation of critical constants only for the special case of one predictor variable. In this article, a method is proposed to construct constant width bands when there are any number of predictor variables. A new criterion for assessing a confidence band is also proposed; it is the probability that a confidence band excludes a false regression function and can be viewed as the power function of a test associated, naturally, with a confidence band. Under this criterion, a numerical comparison between the Scheffé type and constant width bands is then carried out. It emerges from this comparison that the constant width bands can be better than the Scheffé type bands for certain designs.

Original languageEnglish (US)
Pages (from-to)425-436
Number of pages12
JournalJournal of Statistical Computation and Simulation
Volume75
Issue number6
DOIs
StatePublished - Jun 1 2005
Externally publishedYes

Fingerprint

Simultaneous Confidence Bands
Multiple Linear Regression
Confidence Bands
Linear regression
Predictors
Bandwidth
Interval
Regression Function
Statistics
Power Function
Numerical Comparisons
Straight Line
Linear Function
Multiple linear regression
Confidence
Regression

Keywords

  • Linear regression
  • Sensitivity of a confidence band
  • Simultaneous confidence bands
  • Statistical simulation

ASJC Scopus subject areas

  • Statistics and Probability
  • Modeling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

Constant width simultaneous confidence bands in multiple linear regression with predictor variables constrained in intervals. / Liu, W.; Jamshidian, M.; Zhang, Ying; Bretz, F.

In: Journal of Statistical Computation and Simulation, Vol. 75, No. 6, 01.06.2005, p. 425-436.

Research output: Contribution to journalArticle

@article{0b3022410f94470696ad54fa42c2329b,
title = "Constant width simultaneous confidence bands in multiple linear regression with predictor variables constrained in intervals",
abstract = "In the last fifty years, a great deal of research effort has been made on the construction of simultaneous confidence bands for a linear regression function. Two most frequently quoted confidence bands in the statistics literature are the Scheff{\'e} type and constant width bands over a given rectangular region of the predictor variables. For the constant width bands, a method is given by Gafarian [Gafarian, A.V., 1964, Confidence bands in straight line regression. Journal of the American Statistical Association, 59, 182-213.] for the calculation of critical constants only for the special case of one predictor variable. In this article, a method is proposed to construct constant width bands when there are any number of predictor variables. A new criterion for assessing a confidence band is also proposed; it is the probability that a confidence band excludes a false regression function and can be viewed as the power function of a test associated, naturally, with a confidence band. Under this criterion, a numerical comparison between the Scheff{\'e} type and constant width bands is then carried out. It emerges from this comparison that the constant width bands can be better than the Scheff{\'e} type bands for certain designs.",
keywords = "Linear regression, Sensitivity of a confidence band, Simultaneous confidence bands, Statistical simulation",
author = "W. Liu and M. Jamshidian and Ying Zhang and F. Bretz",
year = "2005",
month = "6",
day = "1",
doi = "10.1080/00949650410001701715",
language = "English (US)",
volume = "75",
pages = "425--436",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "6",

}

TY - JOUR

T1 - Constant width simultaneous confidence bands in multiple linear regression with predictor variables constrained in intervals

AU - Liu, W.

AU - Jamshidian, M.

AU - Zhang, Ying

AU - Bretz, F.

PY - 2005/6/1

Y1 - 2005/6/1

N2 - In the last fifty years, a great deal of research effort has been made on the construction of simultaneous confidence bands for a linear regression function. Two most frequently quoted confidence bands in the statistics literature are the Scheffé type and constant width bands over a given rectangular region of the predictor variables. For the constant width bands, a method is given by Gafarian [Gafarian, A.V., 1964, Confidence bands in straight line regression. Journal of the American Statistical Association, 59, 182-213.] for the calculation of critical constants only for the special case of one predictor variable. In this article, a method is proposed to construct constant width bands when there are any number of predictor variables. A new criterion for assessing a confidence band is also proposed; it is the probability that a confidence band excludes a false regression function and can be viewed as the power function of a test associated, naturally, with a confidence band. Under this criterion, a numerical comparison between the Scheffé type and constant width bands is then carried out. It emerges from this comparison that the constant width bands can be better than the Scheffé type bands for certain designs.

AB - In the last fifty years, a great deal of research effort has been made on the construction of simultaneous confidence bands for a linear regression function. Two most frequently quoted confidence bands in the statistics literature are the Scheffé type and constant width bands over a given rectangular region of the predictor variables. For the constant width bands, a method is given by Gafarian [Gafarian, A.V., 1964, Confidence bands in straight line regression. Journal of the American Statistical Association, 59, 182-213.] for the calculation of critical constants only for the special case of one predictor variable. In this article, a method is proposed to construct constant width bands when there are any number of predictor variables. A new criterion for assessing a confidence band is also proposed; it is the probability that a confidence band excludes a false regression function and can be viewed as the power function of a test associated, naturally, with a confidence band. Under this criterion, a numerical comparison between the Scheffé type and constant width bands is then carried out. It emerges from this comparison that the constant width bands can be better than the Scheffé type bands for certain designs.

KW - Linear regression

KW - Sensitivity of a confidence band

KW - Simultaneous confidence bands

KW - Statistical simulation

UR - http://www.scopus.com/inward/record.url?scp=27844573316&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27844573316&partnerID=8YFLogxK

U2 - 10.1080/00949650410001701715

DO - 10.1080/00949650410001701715

M3 - Article

VL - 75

SP - 425

EP - 436

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 6

ER -