The chi-square coding test for nested Markov random field hypotheses
Hardouin, Cécile ; Guyon, Xavier
HAL, hal-00110636 / Harvested from HAL
Let Y be a Markov random foeld on S parametrized via its local conditional specifications. We study consistency of coding and pseudo-likelihood estimators. Then we obtain conditional asymptotic normality for the coding estimator and deduce that the difference of coding statistic for two nested hypotheses is, unconditionally, a chi-square. For these results, we do not need regularity of the lattice, translation invariance for the specification or weak dependence for the field.
Publié le : 1992-07-05
Classification:  Conditional specification,  Markov random field,  coding and pseudo-likelihood estimators,  difference of coding tests,  62M02, 62M05,  [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST]
@article{hal-00110636,
     author = {Hardouin, C\'ecile and Guyon, Xavier},
     title = {The chi-square coding test for nested Markov random field hypotheses},
     journal = {HAL},
     volume = {1992},
     number = {0},
     year = {1992},
     language = {en},
     url = {http://dml.mathdoc.fr/item/hal-00110636}
}
Hardouin, Cécile; Guyon, Xavier. The chi-square coding test for nested Markov random field hypotheses. HAL, Tome 1992 (1992) no. 0, . http://gdmltest.u-ga.fr/item/hal-00110636/