English

Model selection for Poisson processes with covariates

Statistics Theory 2013-06-14 v2 Statistics Theory

Abstract

We observe nn inhomogeneous Poisson processes with covariates and aim at estimating their intensities. We assume that the intensity of each Poisson process is of the form s(,x)s (\cdot, x) where xx is the covariate and where ss is an unknown function. We propose a model selection approach where the models are used to approximate the multivariate function ss. We show that our estimator satisfies an oracle-type inequality under very weak assumptions both on the intensities and the models. By using an Hellinger-type loss, we establish non-asymptotic risk bounds and specify them under several kind of assumptions on the target function ss such as being smooth or a product function. Besides, we show that our estimation procedure is robust with respect to these assumptions.

Keywords

Cite

@article{arxiv.1112.5634,
  title  = {Model selection for Poisson processes with covariates},
  author = {Mathieu Sart},
  journal= {arXiv preprint arXiv:1112.5634},
  year   = {2013}
}
R2 v1 2026-06-21T19:56:30.112Z