Segmentation of the Poisson and negative binomial rate models: a penalized estimator

• DOI: 10.1051/ps/2014005
• Publication Date: 2014-02-25T17:34:10Z
• Authors: Alice Cleynen; Emilie Lebarbier
• Keywords: FOS: Computer and information sciences; model selection; [SDV]Life Sciences [q-bio]; change-point detection; Mathematics - Statistics Theory; Statistics Theory (math.ST); poisson and negative binomial distributions; 01 natural sciences; count data (RNA-seq); 62G05, 62G07, 62P10; [SDV] Life Sciences [q-bio]; Methodology (stat.ME); change-point detection;count data (RNA-seq);poisson and negative binomial distributions;model selection;distribution estimation; distribution estimation; FOS: Mathematics; 0101 mathematics; Statistics - Methodology

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