Student Sliced Inverse Regression

Alessandro Chiancone 1, 2, 3 Florence Forbes 2 Stéphane Girard 2
1 GIPSA-SIGMAPHY - SIGMAPHY
GIPSA-DIS - Département Images et Signal
2 MISTIS - Modelling and Inference of Complex and Structured Stochastic Systems
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : Sliced Inverse Regression (SIR) has been extensively used to reduce the dimension of the predictor space before performing regression. SIR is originally a model free method but it has been shown to actually correspond to the maximum likelihood of an inverse regression model with Gaussian errors. This intrinsic Gaussianity of standard SIR may explain its high sensitivity to outliers as observed in a number of studies. To improve robustness, the inverse regression formulation of SIR is therefore extended to non-Gaussian errors with heavy-tailed distributions. Considering Student distributed errors it is shown that the inverse regression remains tractable via an Expectation- Maximization (EM) algorithm. The algorithm is outlined and tested in the presence of outliers, both in simulated and real data, showing improved results in comparison to a number of other existing approaches.
Document type :
Journal articles
Liste complète des métadonnées

Cited literature [33 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01294982
Contributor : Alessandro Chiancone <>
Submitted on : Monday, August 8, 2016 - 10:10:57 AM
Last modification on : Monday, April 30, 2018 - 3:02:01 PM
Document(s) archivé(s) le : Wednesday, November 9, 2016 - 11:46:45 AM

Files

elsarticle-StudentSIR.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Alessandro Chiancone, Florence Forbes, Stéphane Girard. Student Sliced Inverse Regression. Computational Statistics and Data Analysis, Elsevier, 2017, 113, pp.441-456. ⟨10.1016/j.csda.2016.08.004⟩. ⟨hal-01294982v3⟩

Share

Metrics

Record views

1915

Files downloads

277