We consider the segmentation problem of Poisson and negative binomial (i.e.overdispersed Poisson) rate distributions. In segmentation, an important issueremains the choice of the number of segments. To this end, we propose a penalized-likelihood estimator where the penaltyfunction is constructed in a non-asymptotic context following the works of L. Birgé and P.Massart. The resulting estimator is proved to satisfy an oracle inequality. Theperformances of our criterion is assessed using simulated and real datasets in the RNA-seqdata analysis context.
