The issue of bias in multiple regression is not new. However, researchers are often tempted to ignore the problem, hoping that any biases will have only a small effect on results. The aim of this paper is to draw attention to problems of bias and show that, in some cases, such as feed efficiency testing, biases can be important. Formulae to correct for bias can be derived from the least squares equations. In an example using data from beef cattle in northern New South Wales, Australia, phenotypic regression equations for residual feed efficiency differed from equations derived from genotypic regression and feed standards formulae. This paper investigates whether correcting for bias provides more consistent estimates. 

D. L Robinson

Proceedings of the World Congress on Genetics Applied to Livestock Production, Volume 2002. Session 10, , 10.29, 2002
Download Full PDF BibTEX Citation Endnote Citation Search the Proceedings

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.