Model selection and estimation of a component in additive regression

DOI10.1051/PS/2012028MaRDI QIDQ5174344zbMATH OpenOpenAlexFDO

Authors Xavier Gendre

Publication date 17 February 2015

Published in ESAIM: Probability and Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1209.6534, http://oatao.univ-toulouse.fr/22773/1/Gendre_22773.pdf

zbMATH Keywords

additive regressioncorrelated dataminimax ratemodel selectionnonparametric regressionoracle inequalitypenalized criterion

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08)

Abstract: Let $Yin{R}^n$ be a random vector with mean $s$ and covariance matrix $sigm{a}^2{P}_{n}ra{P}_n$ where $P_n$ is some known $nimesn$-matrix. We construct a statistical procedure to estimate $s$ as well as under moment condition on $Y$ or Gaussian hypothesis. Both cases are developed for known or unknown $sigm{a}^2$. Our approach is free from any prior assumption on $s$ and is based on non-asymptotic model selection methods. Given some linear spaces collection $S_m,minM$, we consider, for any $minM$, the least-squares estimator $hat{s}_m$ of $s$ in $S_m$. Considering a penalty function that is not linear in the dimensions of the $S_m$'s, we select some $hatminM$ in order to get an estimator $hat{s}_{hatm}$ with a quadratic risk as close as possible to the minimal one among the risks of the $hat{s}_m$'s. Non-asymptotic oracle-type inequalities and minimax convergence rates are proved for $hat{s}_{hatm}$. A special attention is given to the estimation of a non-parametric component in additive models. Finally, we carry out a simulation study in order to illustrate the performances of our estimators in practice.

Recommendations

• Model selection for regression on a random design
• Model selection in nonparametric regression
• Selection of components and degrees of smoothing via Lasso in high dimensional nonparametric additive models
• Variable selection in nonparametric additive models
• Component selection in additive quantile regression models

Cited in

(13)

• Additive versus multiplicative models in ecologic regression
• Model selection in nonparametric regression
• Laplace deconvolution with noisy observations
• Gaussian linear model selection in a dependent context
• Variable selection for additive model via cumulative ratios of empirical strengths total
• scientific article; zbMATH DE number 7234919 (Why is no real title available?)
• Laplace deconvolution with dependent errors: a minimax study
• Clarification: Regression Model Selection—A Residual Likelihood Approach
• Effect Selection and Regularization in Structured Additive Distributional Regression
• Estimation in nonparametric regression model with additive and multiplicative noise via Laguerre series
• Additive model selection
• Nonparametric estimation for independent and identically distributed stochastic differential equations with space-time dependent coefficients
• Wavelet estimation for the nonparametric additive model in random design and long-memory dependent errors

This page was built for publication: Model selection and estimation of a component in additive regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5174344)