Application of Bayesian simulation framework in quantitatively measuring presence of competition in living species

This article uses Bayesian simulation algorithms in a checkerboard matrix framework in order to study whether competition can be statistically detected among living species. We study an exhaustive set of binary co-occurrence matrices for habitation data. We categorize the living species into five distinct groups: (1) Mammals; (2) Plants; (3) Birds; (4) Marine Life; and (5) Reptiles. We implement the Holding-swap and Metropolis-swap simulation algorithms to statistically detect the presence of competition for habitation. We find that for ~50% of our dataset, there is statistically significant presence of competition. We observe the following ranking for percentage of dataset with significant level of competition: (1) 90% of birds show competition; (2) 50% of the dataset of reptiles show competition; (3) 40% of mammals and plants; and (4) 20% of the marine life exhibit statistically significant presence of competition. We conclude that birds value habitation more strongly than marine life.