Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs
Annals of the Institute of Statistical Mathematics · 2017-06 · 69 (3) · 673-704
DOI: 10.1007/s10463-016-0560-2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Annals of the Institute of Statistical Mathematics, online erschienen am 05.04.2016. Verfasst von Elizabeth Gross und 2 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 49 Literaturangaben.
Abstract (Originalsprache)
Social networks and other sparse data sets pose significant challenges for statistical inference, since many standard statistical methods for testing model/data fit are not applicable in such settings. Algebraic statistics offers a theoretically justified approach to goodness-of-fit testing that relies on the theory of Markov bases. Most current practices require the computation of the entire basis, which is infeasible in many practical settings. We present a dynamic approach to explore the fiber of a model, which bypasses this issue, and is based on the combinatorics of hypergraphs arising from the toric algebra structure of log-linear models. We demonstrate the approach on the Holland–Leinhardt $$p_1$$p1 model for random directed graphs that allows for reciprocation effects.
Zitieren
APA
Gross, E., Petrović, S., & Stasi, D. (2017). Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs. Annals of the Institute of Statistical Mathematics, 69(3), 673-704. https://doi.org/10.1007/s10463-016-0560-2
Vancouver
Gross E, Petrović S, Stasi D. Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs. Annals of the Institute of Statistical Mathematics. 2017;69(3):673-704. doi:10.1007/s10463-016-0560-2
BibTeX
@article{Gross2017,
author = {Gross, Elizabeth and Petrović, Sonja and Stasi, Despina},
title = {Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs},
journal = {Annals of the Institute of Statistical Mathematics},
year = {2017},
volume = {69},
number = {3},
pages = {673--704},
doi = {10.1007/s10463-016-0560-2}
}
EUVIMED