Maximum likelihood estimation of fossil assemblage composition

Publisher Website
Google Scholar
Abstract
The most common methods of estimating the relative abundance of species in a fossil assemblage are all maximum likelihood estimates. They differ from one another in their inherent assumptions made about the effects of fragmentation and differential preservation in the assemblage. For many fossil assemblages, relative abundance is best estimated by the relative frequency of specimens or relative frequency of elements. Monte Carlo simulations suggest, however, that in most other circumstances estimates based on frequency of elements divided by the number of elements in a complete individual provide greater accuracy than estimates based on minimum number of individuals. This relation results from an interaction between random sampling error and a variety of biases inherent in the two estimates.