1993Communication in Statistics- Theory and MethodsRequires access

R2in restricted linear models

Henning Knautz, Joachim Wilde

Open publisher page 0 citations

Abstract

We consider the coefficient of determination R2 in a linear regression model under linear constraints. We demonstrate that the usual definition of R2 is no longer useful if the restrictions are inhomogeneous and propose a more meaningful alternative.

About this research paper

What this paper is about

We consider the coefficient of determination R2 in a linear regression model under linear constraints. We demonstrate that the usual definition of R2 is no longer useful if the restrictions are inhomogeneous and propose a more meaningful alternative.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

We consider the coefficient of determination R2 in a linear regression model under linear constraints. We demonstrate that the usual definition of R2 is no longer useful if the restrictions are inhomogeneous and propose a more meaningful alternative.

Key concepts: Linear regression, Linear model, Proper linear model, Linear predictor function, Mathematics, Log-linear model, General linear model, Applied mathematics

Related papers

Back to paper searchBrowse research topicsOriginal source
R2in restricted linear models — Research Paper | ScholarLens