2006/05/10 by Manuel Mendoza, Christine M. Janis, Paul Palmqvist · 106 citations
Agricultural and Biological Sciences · Earth and Planetary Sciences · Environmental Science · Mathematics · #Bat Biology and Ecology Studies #Biology #Ecology #Evolution and Paleontology Studies #Mathematics #Regression #Regression analysis #Statistics #Wildlife Ecology and Conservation #Zoology
paper · pdf · doi:10.1111/j.1469-7998.2006.00094.x
published in Journal of Zoology 270(1), 90-101 (Wiley)
openalex publication_date 2006/05/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/15
Abstract The correlation between body mass and both skeletal and dental measures in living mammals has enabled paleontologists to obtain reliable estimates of body size for extinct species, usually using log‐transformed bivariate least‐squares regression equations. Multiple regression, however, has rarely been used for estimating the mass of extinct species, although this technique can clearly improve the predictive equations compared with those adjusted by simple regression. However, the use of multiple regression is problematical, because even those functions explaining a high percentage of the variance of the dependent variable (i.e. body mass) can show a rather limited predictive power. After analyzing which factors determine the predictive ability of multiple regression equations, we propose a new set of algorithms that allow the estimation of the body mass of extinct ungulates. These algorithms are finally applied to three Miocene ungulate species, Dinohippus leidyanus , Stenomylus hitchcocki and Aletomeryx scotti .