B0983
Title: Algebraic model invariants for compositional data
Authors: Eva Riccomagno - Università degli Studi di Genova (Italy)
Orlando Marigliano - University of Genova (Italy) [presenting]
Abstract: Compositional data arise whenever we record the relative quantities of the parts that form a whole. For instance, the proportions of the different cell types in a sample of organic tissue would constitute compositional data. In this talk, I present examples of probability distributions for compositional data and describe algebraic statistical models for these distributions. I also discuss the maximum likelihood degree of these models.
