Geodesic
Large deviations and full edgeworth expansions for finite Markov chains with applications to the analysis of genomic sequences
ESAIM: Probability and Statistics, Tome 14 (2010), pp. 435-455

Voir la notice de l'article provenant de la source Numdam

To establish lists of words with unexpected frequencies in long sequences, for instance in a molecular biology context, one needs to quantify the exceptionality of families of word frequencies in random sequences. To this aim, we study large deviation probabilities of multidimensional word counts for Markov and hidden Markov models. More specifically, we compute local Edgeworth expansions of arbitrary degrees for multivariate partial sums of lattice valued functionals of finite Markov chains. This yields sharp approximations of the associated large deviation probabilities. We also provide detailed simulations. These exhibit in particular previously unreported periodic oscillations, for which we provide theoretical explanations.

DOI : 10.1051/ps/2009008
Classification : 60J10, 60F10, 60J55, 92D20, 60F05
Keywords: Markov chains, hidden Markov models, large deviations, edgeworth expansions, protein and DNA sequences 
@article{PS_2010__14__435_0,
     author = {Pudlo, Pierre},
     title = {Large deviations and full edgeworth expansions for finite {Markov} chains with applications to the analysis of genomic sequences},
     journal = {ESAIM: Probability and Statistics},
     pages = {435--455},
     publisher = {EDP-Sciences},
     volume = {14},
     year = {2010},
     doi = {10.1051/ps/2009008},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.1051/ps/2009008/}
}
TY  - JOUR
AU  - Pudlo, Pierre
TI  - Large deviations and full edgeworth expansions for finite Markov chains with applications to the analysis of genomic sequences
JO  - ESAIM: Probability and Statistics
PY  - 2010
SP  - 435
EP  - 455
VL  - 14
PB  - EDP-Sciences
UR  - http://geodesic.mathdoc.fr/articles/10.1051/ps/2009008/
DO  - 10.1051/ps/2009008
LA  - en
ID  - PS_2010__14__435_0
ER  - 
%0 Journal Article
%A Pudlo, Pierre
%T Large deviations and full edgeworth expansions for finite Markov chains with applications to the analysis of genomic sequences
%J ESAIM: Probability and Statistics
%D 2010
%P 435-455
%V 14
%I EDP-Sciences
%U http://geodesic.mathdoc.fr/articles/10.1051/ps/2009008/
%R 10.1051/ps/2009008
%G en
%F PS_2010__14__435_0
Pudlo, Pierre. Large deviations and full edgeworth expansions for finite Markov chains with applications to the analysis of genomic sequences. ESAIM: Probability and Statistics, Tome 14 (2010), pp. 435-455. doi: 10.1051/ps/2009008

Cité par Sources :