As more and more network-structured data sets are available, the statistical
analysis of valued graphs has become common place. Looking for a latent
structure is one of the many strategies used to better understand the behavior
of a network. Several methods already exist for the binary case. We present a
model-based strategy to uncover groups of nodes in valued graphs. This
framework can be used for a wide span of parametric random graphs models and
allows to include covariates. Variational tools allow us to achieve approximate
maximum likelihood estimation of the parameters of these models. We provide a
simulation study showing that our estimation method performs well over a broad
range of situations. We apply this method to analyze host--parasite interaction
networks in forest ecosystems.Comment: Published in at http://dx.doi.org/10.1214/10-AOAS361 the Annals of
Applied Statistics (http://www.imstat.org/aoas/) by the Institute of
Mathematical Statistics (http://www.imstat.org
Summary1. Pest regulation is an important ecosystem service provided by biodiversity, as plants growing in species-rich communities often experience associational resistance to herbivores. However, little is known about the respective influence of the quantity and identity of associated species on herbivory in focal plants.2. Using a meta-analysis to compare insect herbivory in pure and mixed forests, we specifically tested the effects of the relative abundance of focal tree species and of phylogenetic distance between focal and associated tree species on the magnitude of associational resistance. 3. Overall, insect herbivory was significantly lower in mixed forests, but the outcome varied greatly depending on the phylogenetic relatedness among tree species and the degree of herbivore feeding specialization. 4. Specialist herbivore damage or abundance was positively related to relative abundance of their host trees, regardless of the phylogenetic distance between host and associated tree species. 5. By contrast, tree diversity triggered associational resistance to generalist herbivores only when tree mixtures included tree species phylogenetically distant to the focal species. 6. Synthesis and applications. Our study demonstrates that the establishment of mixed forests per se is not sufficient to convey associational resistance to herbivores if the identity of tree species associated in mixtures is not taken into account. As a general rule, mixing phylogenetically more distinct tree species, such as mixtures of conifers and broadleaved trees, results in more effective reduction in herbivore damage.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.