The Annals of Statistics

Accounting for Intrinsic Nonlinearity in Nonlinear Regression Parameter Inference Regions

David C. Hamilton, Donald G. Watts, and Douglas M. Bates

Full-text: Open access

Abstract

Joint confidence and likelihood regions for the parameters in nonlinear regression models can be defined using the geometric concepts of sample space and solution locus. Using a quadratic approximation to the solution locus, instead of the usual linear approximation, it is shown that these inference regions correspond to ellipsoids on the tangent plane at the least squares point. Accurate approximate inference regions can be obtained by mapping these ellipsoids into the parameter space, and measures of the effect of intrinsic nonlinearity on inference can be based on the difference between the tangent plane ellipsoids and the sphere which would be obtained using a linear approximation.

Article information

Source
Ann. Statist., Volume 10, Number 2 (1982), 386-393.

Dates
First available in Project Euclid: 12 April 2007

Permanent link to this document
https://projecteuclid.org/euclid.aos/1176345780

Digital Object Identifier
doi:10.1214/aos/1176345780

Mathematical Reviews number (MathSciNet)
MR653514

Zentralblatt MATH identifier
0537.62045

JSTOR
links.jstor.org

Subjects
Primary: 62J02: General nonlinear regression
Secondary: 62F25: Tolerance and confidence regions

Keywords
Intrinsic curvature nonlinear regression approximate inference regions

Citation

Hamilton, David C.; Watts, Donald G.; Bates, Douglas M. Accounting for Intrinsic Nonlinearity in Nonlinear Regression Parameter Inference Regions. Ann. Statist. 10 (1982), no. 2, 386--393. doi:10.1214/aos/1176345780. https://projecteuclid.org/euclid.aos/1176345780


Export citation