We propose a new empirical procedure for detecting phase transition
from a single sample of a Gibbs-Markov random field. The method is based on
frequencies for large deviations when the whole sample is divided in smaller
blocks and estimates for the rate function. We relate our approach to an almost
sure large deviation principle.
Publié le : 1997-05-14
Classification:
Gibbs measures,
Markov random fields,
phase transition,
large deviations,
60G60,
60F10,
62M30,
82B26
@article{1034625344,
author = {Comets, Francis},
title = {Detecting phase transition for Gibbs measures},
journal = {Ann. Appl. Probab.},
volume = {7},
number = {1},
year = {1997},
pages = { 545-563},
language = {en},
url = {http://dml.mathdoc.fr/item/1034625344}
}
Comets, Francis. Detecting phase transition for Gibbs measures. Ann. Appl. Probab., Tome 7 (1997) no. 1, pp. 545-563. http://gdmltest.u-ga.fr/item/1034625344/