Polyhedral conditions for the nonexistence of the MLE for hierarchical log-linear models
Nicholas Eriksson, Stephen E. Fienberg, Alessandro Rinaldo, Seth Sullivant
Abstract
We provide a polyhedral description of the conditions for the existence of the maximum likelihood estimate (MLE) for a hierarchical log-linear model. The MLE exists if and only if the observed margins lie in the relative interior of the marginal cone. Using this description, we give an algorithm for determining if the MLE exists. If the tree width is bounded, the algorithm runs in polynomial time. We also perform a computational study of the case of three random variables under the no three-factor effect model.
Create a lesson
Related papers
Off-Diagonal Induced Ramsey Numbers
Bruno Andrades, Marcelo Campos, Rafael Filipe et al.
Largest Sperner families with bounded VC-dimension
Gennian Ge, Hanyu Wang, Zixiang Xu et al.
Proof of two conjectures on the log-concavity of partition ranks and cranks
Jianxi Mao, Wenle Shi, Bao-Xuan Zhu
The Exact Number of Pentagons in Book-Free Graphs
Wentao Hu
A family of symmetric counterexamples to the bent sum decomposition conjecture
Máté Róbert Kepes, Anita Kercsó-Molnár, Viczián András
Counting graphs admitting a strictly matched involution
Stephan Dominique Andres, Igor Režan