Model selection for Poisson processes

DOI10.1214/074921707000000265MaRDI QIDQ5324530zbMATH OpenOpenAlexFDO

Authors L. Birgé

Publication date 3 August 2009

Published in Institute of Mathematical Statistics Lecture Notes - Monograph Series (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0609549

zbMATH Keywords

adaptive estimationaggregationintensity estimationmodel selectionrobust tests

Mathematics Subject Classification ID

Approximation by arbitrary linear expressions (41A45) Approximation by arbitrary nonlinear expressions; widths and entropy (41A46) Nonparametric estimation (62G05) Density estimation (62G07) Asymptotic properties of nonparametric inference (62G20) Linear regression; mixed models (62J05) Markov processes: estimation; hidden Markov models (62M05) Non-Markovian processes: estimation (62M09)

Abstract: Our purpose in this paper is to apply the general methodology for model selection based on T-estimators developed in Birg'{e} [Ann. Inst. H. Poincar'{e} Probab. Statist. 42 (2006) 273--325] to the particular situation of the estimation of the unknown mean measure of a Poisson process. We introduce a Hellinger type distance between finite positive measures to serve as our loss function and we build suitable tests between balls (with respect to this distance) in the set of mean measures. As a consequence of the existence of such tests, given a suitable family of approximating models, we can build T-estimators for the mean measure based on this family of models and analyze their performances. We provide a number of applications to adaptive intensity estimation when the square root of the intensity belongs to various smoothness classes. We also give a method for aggregation of preliminary estimators.

Recommendations

• Model selection for Poisson processes with covariates
• Estimator selection with respect to Hellinger-type risks
• Model selection via testing: an alternative to (penalized) maximum likelihood estimators.
• Estimating the intensity of a random measure by histogram type estimators
• Adaptive estimation of the intensity of inhomogeneous Poisson processes via concentration inequalities

Cited in

(16)

• Estimating the intensity of a random measure by histogram type estimators
• Estimating linear functionals of a sparse family of Poisson means
• Near optimal thresholding estimation of a Poisson intensity on the real line
• Optimal choice of observation window for Poisson observations
• Aggregating estimates by convex optimization
• Non-parametric Poisson regression from independent and weakly dependent observations by model selection
• Nonparametric intensity estimation from noisy observations of a Poisson process under unknown error distribution
• Histogram selection for possibly censored data
• Model selection for density estimation with \(\mathbb L_2\)-loss
• Model selection for Poisson processes with covariates
• Robust estimation on a parametric model via testing
• Estimation of the transition density of a Markov chain
• Estimating the conditional density by histogram type estimators and model selection
• Can one hear the shape of a population history?
• Concentration inequalities for Poisson point processes with application to adaptive intensity estimation
• Estimator selection with respect to Hellinger-type risks

This page was built for publication: Model selection for Poisson processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5324530)