Prediction with high dimensional regression via hierarchically structured Gaussian mixtures and latent variables

Chun-Chen Tu 1 Florence Forbes 2 Benjamin Lemasson 3 Naisyin Wang 1
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
3 Equipe 5 : NeuroImagerie Fonctionnelle et Perfusion Cérébrale
UJF - Université Joseph Fourier - Grenoble 1, CEA - Commissariat à l'énergie atomique et aux énergies alternatives, INSERM - Institut National de la Santé et de la Recherche Médicale : U836, [GIN] Grenoble Institut des Neurosciences
Abstract : We propose a hierarchical Gaussian locally linear mapping structured mixture model, named HGLLiM, to predict low dimensional responses based on high dimensional covariates when the associations between the responses and the covariates are non-linear. For tractability, HGLLiM adopts inverse regression to handle the high dimension and locally linear mappings to capture potentially non-linear relations. Data with similar associations are grouped together to form a cluster. A mixture is composed of several clusters following a hierarchical structure. This structure enables shared covariance matrices and latent factors across smaller clusters to limit the number of parameters to estimate. Moreover, HGLLiM adopts a robust estimation procedure for model stability. We use three real data sets to demonstrate different features of HGLLiM. With the face data set, HGLLiM shows ability to model non-linear relationships through mixtures. With the orange juice data set, we show that the prediction performance of HGLLiM is robust to the presence of outliers. Moreover, we demonstrate that HGLLiM is capable of handling large-scale complex data by using the data acquired from a magnetic resonance vascular fingerprinting study. These examples illustrate the wide applicability of HGLLiM to handle different aspects of a complex data structure in prediction.
Document type :
Journal articles
Complete list of metadatas

Cited literature [23 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02263144
Contributor : Florence Forbes <>
Submitted on : Monday, August 12, 2019 - 1:48:12 PM
Last modification on : Monday, August 19, 2019 - 1:07:47 PM

File

HGLLiM_main.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Chun-Chen Tu, Florence Forbes, Benjamin Lemasson, Naisyin Wang. Prediction with high dimensional regression via hierarchically structured Gaussian mixtures and latent variables. Journal of the Royal Statistical Society: Series C Applied Statistics, Wiley, 2019, ⟨10.1111/rssc.12370⟩. ⟨hal-02263144⟩

Share

Metrics

Record views

49

Files downloads

9