Information perspectives of the Haseman-Elston method.

The Haseman-Elston regression approach can be viewed as a special case in a larger framework in which independent sibling pairs are ascertained, perhaps based on their phenotypes. Using likelihood arguments in an idealized setting, expressions are obtained for the inherent linkage information contained in the phenotype distributions. The results provide insight and analytic results that apply to general phenotype distributions and can be used to judge the effects of various data transformations and selective sampling of extreme phenotypes. Copyright 2003 S. Karger AG, Basel