Characterizing the domain of a regression model

Daniel G. Brooks, Steven S. Carroll, William A. Verdini

Research output: Contribution to journalArticle

19 Scopus citations

Abstract

Several frequently used estimates of the domain of a regression model are discussed and compared with the convex hull of the predictor-variable values. A simple procedure for identifying new prediction values as interpolation or extrapolation points is presented using the convex hull as the model domain estimate and is illustrated by examples.

Original languageEnglish (US)
Pages (from-to)187-190
Number of pages4
JournalAmerican Statistician
Volume42
Issue number3
DOIs
StatePublished - Aug 1988

    Fingerprint

Keywords

  • Convex hull
  • Extrapolation
  • Interpolation
  • Model domain

ASJC Scopus subject areas

  • Statistics and Probability
  • Mathematics(all)
  • Statistics, Probability and Uncertainty

Cite this

Brooks, D. G., Carroll, S. S., & Verdini, W. A. (1988). Characterizing the domain of a regression model. American Statistician, 42(3), 187-190. https://doi.org/10.1080/00031305.1988.10475559