Goodness-of-fit tests for regression models

Romain Azaïs 1, 2 Sandie Ferrigno 1, 2 Marie-José Martinez 3
1 BIGS - Biology, genetics and statistics
Inria Nancy - Grand Est, IECL - Institut Élie Cartan de Lorraine
3 SVH - Statistique pour le Vivant et l’Homme
LJK - Laboratoire Jean Kuntzmann
Abstract : The objective is to construct a tool to test the validity of a regression model. For this, we have studied some omnibus tests of goodness-of-fit. Our study focuses on the regression function in the regression model. These tests can be ``directional'' in that they are designed to detect departures from mainly one given assumption of the model (regression function, variance function or error) or global with the conditional distribution function. We have compared, through simulations, different nonparametric methods to test the validity of the model. Two methods are directional and one is global. The establishment of such statistical tests require nonparametric estimators and the use of wild bootstrap methods for the simulations.
Document type :
Poster communications
Liste complète des métadonnées

https://hal.inria.fr/hal-01557315
Contributor : Sandie Ferrigno <>
Submitted on : Thursday, July 6, 2017 - 9:44:59 AM
Last modification on : Friday, October 5, 2018 - 2:28:41 PM

Identifiers

  • HAL Id : hal-01557315, version 1

Citation

Romain Azaïs, Sandie Ferrigno, Marie-José Martinez. Goodness-of-fit tests for regression models. CMStatistics 2017, Dec 2017, London, United Kingdom. ⟨hal-01557315⟩

Share

Metrics

Record views

334