Context Tree Selection: A Unifying View
Garivier, Aurélien ; Leonardi, Florencia
arXiv, 1011.2424 / Harvested from arXiv
The present paper investigates non-asymptotic properties of two popular procedures of context tree (or Variable Length Markov Chains) estimation: Rissanen's algorithm Context and the Penalized Maximum Likelihood criterion. First showing how they are related, we prove finite horizon bounds for the probability of over- and under-estimation. Concerning overestimation, no boundedness or loss-of-memory conditions are required: the proof relies on new deviation inequalities for empirical probabilities of independent interest. The underestimation properties rely on loss-of-memory and separation conditions of the process. These results improve and generalize the bounds obtained previously. Context tree models have been introduced by Rissanen as a parsimonious generalization of Markov models. Since then, they have been widely used in applied probability and statistics.
Publié le : 2010-11-10
Classification:  Mathematics - Statistics Theory,  Mathematics - Probability
@article{1011.2424,
     author = {Garivier, Aur\'elien and Leonardi, Florencia},
     title = {Context Tree Selection: A Unifying View},
     journal = {arXiv},
     volume = {2010},
     number = {0},
     year = {2010},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1011.2424}
}
Garivier, Aurélien; Leonardi, Florencia. Context Tree Selection: A Unifying View. arXiv, Tome 2010 (2010) no. 0, . http://gdmltest.u-ga.fr/item/1011.2424/