Counting co-occurrence matrices and its application to community ecology

Yung-Pin Chen
Department of Mathematical Sciences, Lewis and Clark College

Abstract: Ecologists use co-occurrence matrices to record the presence or absence of certain species in different habitats. They often ask what patterns to expect if the species distributed themselves randomly. This requires counting the number of distinct co-occurrence matrices with fixed row and column sums. I shall introduce and discuss some combinatorial and statistical ideas to that quest. In particular, I will describe how to use Markov chain Monte Carlo to approach this counting problem.