Single index Fréchet regression

DOI10.1214/23-AOS2307arXiv2108.05437OpenAlexW3191092502MaRDI QIDQ6183758FDOQ6183758

Authors: Satarupa Bhattacharjee, Hans-Georg Müller 

Publication date: 4 January 2024

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: Single index models provide an effective dimension reduction tool in regression, especially for high dimensional data, by projecting a general multivariate predictor onto a direction vector. We propose a novel single-index model for regression models where metric space-valued random object responses are coupled with multivariate Euclidean predictors. The responses in this regression model include complex, non-Euclidean data that lie in abstract metric spaces, including covariance matrices, graph Laplacians of networks, and univariate probability distribution functions. While Fr'echet regression has proved useful for modeling the conditional mean of such random objects given multivariate Euclidean vectors, it does not provide for regression parameters such as slopes or intercepts, since the metric space-valued responses are not amenable to linear operations. As a consequence, distributional results for Fr'echet regression have been elusive. We show here that for the case of multivariate Euclidean predictors, the parameters that define a single index and projection vector can be used to substitute for the inherent absence of parameters in Fr'echet regression. Specifically, we derive the asymptotic distribution of suitable estimates of these parameters, which then can be utilized to test linear hypotheses for the parameters, subject to an identifiability condition. We demonstrate the finite sample performance of estimation and inference for the proposed single index Fr'echet regression model through simulation studies. The method is illustrated for resting-state functional Magnetic Resonance Imaging (fMRI) data from the ADNI study.

Full work available at URL: https://arxiv.org/abs/2108.05437

Recommendations

• Fréchet regression for random objects with Euclidean predictors
• Fréchet kernel sliced inverse regression
• Estimation and inference in functional single-index models
• Total variation regularized Fréchet regression for metric-space valued data
• A functional single-index model

zbMATH Keywords

dimension reductionrandom objectsinferenceM-estimatornon-Euclidean data

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Nonparametric hypothesis testing (62G10) Asymptotic properties of nonparametric inference (62G20) Statistics on metric spaces (62R20)

Cites Work

• A constructive approach to the estimation of dimension reduction directions
• Title not available (Why is that?)
• Likelihood-based local polynomial fitting for single-index models
• Optimal smoothing in single-index models
• Semi-parametric estimation of partially linear single-index models
• Sliced Inverse Regression for Dimension Reduction
• Generalized Partially Linear Single-Index Models
• Penalized Spline Estimation for Partially Linear Single-Index Models
• Estimation and testing for partially linear single-index models
• Semiparametric least squares (SLS) and weighted SLS estimation of single-index models
• Fréchet regression for random objects with Euclidean predictors
• Investigating Smooth Multiple Regression by the Method of Average Derivatives
• Consistent Estimation of Scaled Coefficients
• Title not available (Why is that?)
• Profile likelihood inferences on semiparametric varying-coefficient partially linear models
• On Directional Regression for Dimension Reduction
• The jackknife and bootstrap
• Estimating the asymptotic covariance matrix for quantile regression models. A Monte Carlo study
• Functional single index models for longitudinal data
• Identifiability of single-index models and additive-index models
• Total variation regularized Fréchet regression for metric-space valued data
• Uniform convergence of local Fréchet regression with applications to locating extrema and time warping for metric space valued trajectories
• A course in metric geometry
• Nonparametric statistics on manifolds and their applications to object data analysis
• Overview of object oriented data analysis
• Riemannian \(L^{p}\) center of mass: existence, uniqueness, and convexity
• Fréchet single index models for object response regression
• Estimation for a partial-linear single-index model
• Weak and strong uniform consistency of the kernel estimate of a density and its derivatives
• Single and multiple index functional regression models with nonparametric link
• On distribution-weighted partial least squares with diverging number of highly correlated predictors
• Empirical Likelihood Confidence Regions in a Partially Linear Single-Index Model
• The EFM approach for single-index models
• Large margin methods for structured and interdependent output variables
• Large sample theory of intrinsic and extrinsic sample means on manifolds. II.
• The statistical theory of shape
• Single index regression models in the presence of censoring depending on the covariates
• On an asymptotically more efficient estimation of the single-index model
• Peter Hall, functional data analysis and random objects
• Dimension reduction based on constrained canonical correlation and variable filtering
• Bootstrap Standard Error Estimates for Linear Regression
• Quantile regression for the single-index coefficient model
• A geometric study of Wasserstein spaces: Euclidean spaces
• Barycentric subspace analysis on manifolds
• Principal single-index varying-coefficient models for dimension reduction in quantile regression
• A note on moment convergence of bootstrap M-estimators
• Convergence rates for the generalized Fréchet mean via the quadruple inequality
• Efficient estimation in single index models through smoothing splines
• Automatic and location-adaptive estimation in functional single-index regression
• Fréchet estimation of time-varying covariance matrices from sparse data, with application to the regional co-evolution of myelination in the developing brain
• Wasserstein \(F\)-tests and confidence bands for the Fréchet regression of density response curves
• Concurrent object regression
• Choosing the most relevant level sets for depicting a sample of densities
• Nonlinear generalization of the monotone single index model
• Weak convergence and empirical processes. With applications to statistics
• Fréchet sufficient dimension reduction for random objects

Cited In (1)

• Metric statistics: exploration and inference for random objects with distance profiles

This page was built for publication: Single index Fréchet regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6183758)