Local independence feature screening for nonparametric and semiparametric models by marginal empirical likelihood

DOI10.1214/15-AOS1374zbMATH Open1486.62082arXiv1502.07061WikidataQ39723923 ScholiaQ39723923MaRDI QIDQ282446FDOQ282446

Authors: Jinyuan Chang, Cheng Yong Tang, Yichao Wu 

Publication date: 12 May 2016

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

Abstract: We consider an independence feature screening technique for identifying explanatory variables that locally contribute to the response variable in high-dimensional regression analysis. Without requiring a specific parametric form of the underlying data model, our approach accommodates a wide spectrum of nonparametric and semiparametric model families. To detect the local contributions of explanatory variables, our approach constructs empirical likelihood locally in conjunction with marginal nonparametric regressions. Since our approach actually requires no estimation, it is advantageous in scenarios such as the single-index models where even specification and identification of a marginal model is an issue. By automatically incorporating the level of variation of the nonparametric regression and directly assessing the strength of data evidence supporting local contribution from each explanatory variable, our approach provides a unique perspective for solving feature screening problems. Theoretical analysis shows that our approach can handle data dimensionality growing exponentially with the sample size. With extensive theoretical illustrations and numerical examples, we show that the local independence screening approach performs promisingly.

Full work available at URL: https://arxiv.org/abs/1502.07061

Recommendations

• Feature screening for nonparametric and semiparametric models with ultrahigh-dimensional covariates
• Robust conditional nonparametric independence screening for ultrahigh-dimensional data
• Marginal empirical likelihood and sure independence feature screening
• Nonparametric independence screening in sparse ultra-high-dimensional additive models
• Nonparametric Independence Screening in Sparse Ultra-High-Dimensional Varying Coefficient Models

zbMATH Keywords

empirical likelihoodhigh-dimensional data analysisnonparametric and semiparametric modelssure independence screening

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20) Estimation in multivariate analysis (62H12)

Cites Work

• Sure independence screening in generalized linear models with NP-dimensionality
• Model-free feature screening for ultrahigh-dimensional data
• Empirical likelihood confidence intervals for nonparametric density estimation
• Statistics for high-dimensional data. Methods, theory and applications.
• Title not available (Why is that?)
• Feature screening via distance correlation learning
• Quadratic Inference Functions for Varying‐Coefficient Models with Longitudinal Data
• Extending the scope of empirical likelihood
• Additive regression and other nonparametric models
• Empirical likelihood for linear models
• Empirical likelihood
• Penalized high-dimensional empirical likelihood
• A Selective Overview of Variable Selection in High Dimensional Feature Space (Invited Review Article)
• Empirical likelihood ratio confidence intervals for a single functional
• Title not available (Why is that?)
• Feature selection for varying coefficient models with ultrahigh-dimensional covariates
• Nonparametric independence screening in sparse ultra-high-dimensional additive models
• On varying-coefficient independence screening for high-dimensional varying-coefficient models
• Nonparametric Independence Screening in Sparse Ultra-High-Dimensional Varying Coefficient Models
• Component selection and smoothing in multivariate nonparametric regression
• Variable selection in nonparametric additive models
• Title not available (Why is that?)
• Varying-coefficient models and basis function approximations for the analysis of repeated measurements
• Effects of data dimension on empirical likelihood
• A review on empirical likelihood methods for regression
• Ultrahigh dimensional feature selection: beyond the linear model
• Penalized empirical likelihood and growing dimensional general estimating equations
• Weighted Local Regression and Kernel Methods for Nonparametric Curve Fitting
• High dimensional generalized empirical likelihood for moment restrictions with dependent data
• Empirical likelihood confidence intervals for local linear smoothers
• Local independence feature screening for nonparametric and semiparametric models by marginal empirical likelihood
• Title not available (Why is that?)
• Marginal empirical likelihood and sure independence feature screening
• The Kolmogorov filter for variable screening in high-dimensional binary classification
• Robust rank correlation based screening

Cited In (25)

• A model-averaging method for high-dimensional regression with missing responses at random
• A model-free conditional screening approach via sufficient dimension reduction
• Local independence feature screening for nonparametric and semiparametric models by marginal empirical likelihood
• A new scope of penalized empirical likelihood with high-dimensional estimating equations
• Marginal empirical likelihood and sure independence feature screening
• Copula-based Partial Correlation Screening: a Joint and Robust Approach
• Nonparametric independence screening via favored smoothing bandwidth
• Covariate selection under nonignorable nonresponse
• Coordinatewise Gaussianization: Theories and Applications
• Variable importance assessments and backward variable selection for multi-sample problems
• A general framework for tensor screening through smoothing
• Sufficient variable selection using independence measures for continuous response
• Conditional feature screening for mean and variance functions in models with multiple-index structure
• A nonparametric procedure for linear and nonlinear variable screening
• Ultrahigh dimensional feature screening for additive model with multivariate response
• On Sure Screening with Multiple Responses
• Self-normalization: taming a wild population in a heavy-tailed world
• Balanced augmented empirical likelihood for regression models
• Robust Feature Screening via Distance Correlation for Ultrahigh Dimensional Data With Responses Missing at Random
• Independence index sufficient variable screening for categorical responses
• Conditional sure independence screening by conditional marginal empirical likelihood
• Refined Cramér-type moderate deviation theorems for general self-normalized sums with applications to dependent random variables and winsorized mean
• Empirical likelihood ratio tests for non-nested model selection based on predictive losses
• Conditional SIRS for nonparametric and semiparametric models by marginal empirical likelihood
• Wilks' theorem for semiparametric regressions with weakly dependent data

Uses Software

• ElemStatLearn

This page was built for publication: Local independence feature screening for nonparametric and semiparametric models by marginal empirical likelihood

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q282446)