EUVIMEDObjevte EUVIMED
Crossref

Accounting for Missing Actors in Interaction Network Inference from Abundance Data

Raphaëlle Momal, Stéphane Robin, Christophe Ambroise

Journal of the Royal Statistical Society Series C: Applied Statistics · 2021-11-01 · 70 (5) · 1230-1258

DOI: 10.1111/rssc.12509 ↗

At a glance

Journal article in Journal of the Royal Statistical Society Series C: Applied Statistics, published online on 2021-11-17. Written by Raphaëlle Momal and 2 co-authors. Cited 6 times so far (Crossref); 53 references.

AbstraktBibliografické údajeCiteRelated studies

Abstract (original language)

Network inference aims at unravelling the dependency structure relating jointly observed variables. Graphical models provide a general framework to distinguish between marginal and conditional dependency. Unobserved variables (missing actors) may induce apparent conditional dependencies. In the context of count data, we introduce a mixture of Poisson log-normal distributions with tree-shaped graphical models, to recover the dependency structure, including missing actors. We design a variational EM algorithm and assess its performance on synthetic data. We demonstrate the ability of our approach to recover environmental drivers on two ecological data sets. The corresponding R package is available from github.com/Rmomal/nestor.

Cite

APA

Momal, R., Robin, S., & Ambroise, C. (2021). Accounting for Missing Actors in Interaction Network Inference from Abundance Data. Journal of the Royal Statistical Society Series C: Applied Statistics, 70(5), 1230-1258. https://doi.org/10.1111/rssc.12509

Vancouver

Momal R, Robin S, Ambroise C. Accounting for Missing Actors in Interaction Network Inference from Abundance Data. Journal of the Royal Statistical Society Series C: Applied Statistics. 2021;70(5):1230-1258. doi:10.1111/rssc.12509

BibTeX
@article{Momal2021,
  author = {Momal, Raphaëlle and Robin, Stéphane and Ambroise, Christophe},
  title = {Accounting for Missing Actors in Interaction Network Inference from Abundance Data},
  journal = {Journal of the Royal Statistical Society Series C: Applied Statistics},
  year = {2021},
  volume = {70},
  number = {5},
  pages = {1230--1258},
  doi = {10.1111/rssc.12509}
}
Start for free →