Sure independence screening in generalized linear models with NP-dimensionality

DOI10.1214/10-AOS798MaRDI QIDQ140975zbMATH OpenOpenAlexFDO

Authors Rui Song, Jianqing Fan

Publication date 1 December 2010

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

Full work available at URL https://arxiv.org/abs/0903.5255

zbMATH Keywords

generalized linear modelsindependent learningsure independent screeningvariable selection

Mathematics Subject Classification ID

Large deviations (60F10) Statistical distribution theory (62E99) Generalized linear models (logistic models) (62J12) Computational learning theory (68Q32)

Abstract: Ultrahigh-dimensional variable selection plays an increasingly important role in contemporary scientific discoveries and statistical research. Among others, Fan and Lv [J. R. Stat. Soc. Ser. B Stat. Methodol. 70 (2008) 849-911] propose an independent screening framework by ranking the marginal correlations. They showed that the correlation ranking procedure possesses a sure independence screening property within the context of the linear model with Gaussian covariates and responses. In this paper, we propose a more general version of the independent learning with ranking the maximum marginal likelihood estimates or the maximum marginal likelihood itself in generalized linear models. We show that the proposed methods, with Fan and Lv [J. R. Stat. Soc. Ser. B Stat. Methodol. 70 (2008) 849-911] as a very special case, also possess the sure screening property with vanishing false selection rate. The conditions under which the independence learning possesses a sure screening is surprisingly simple. This justifies the applicability of such a simple method in a wide spectrum. We quantify explicitly the extent to which the dimensionality can be reduced by independence screening, which depends on the interactions of the covariance matrix of covariates and true parameters. Simulation studies are used to illustrate the utility of the proposed approaches. In addition, we establish an exponential inequality for the quasi-maximum likelihood estimator which is useful for high-dimensional statistical learning.

Recommendations

• Robust sure independence screening for nonpolynomial dimensional generalized linear models
• Sure independence screening in ultrahigh dimensional generalized additive models
• A sure independence screening procedure for ultra-high dimensional partially linear additive models
• Sure independence screening adjusted for confounding covariates with ultrahigh dimensional data
• Robust sure independence screening for ultrahigh dimensional non-normal data
• Nonparametric Independence Screening in Sparse Ultra-High-Dimensional Varying Coefficient Models
• Nonparametric independence screening in sparse ultra-high-dimensional additive models
• Independent screening in high-dimensional exponential family predictors’ space
• Principled sure independence screening for Cox models with ultra-high-dimensional covariates
• Nonparametric independence screening for ultra-high dimensional generalized varying coefficient models with longitudinal data

Cites work

• A Statistical View of Some Chemometrics Regression Tools
• About the constants in Talagrand's concentration inequalities for empirical processes.
• An Analysis of Transformations Revisited
• Asymptotic properties of bridge estimators in sparse high-dimensional regression models
• Consistency and asymptotic normality of the maximum likelihood estimator in generalized linear models
• High-dimensional classification using features annealed independence rules
• High-dimensional generalized linear models and the lasso
• scientific article; zbMATH DE number 49190 (Why is no real title available?)
• scientific article; zbMATH DE number 3626409 (Why is no real title available?)
• scientific article; zbMATH DE number 845714 (Why is no real title available?)
• scientific article; zbMATH DE number 3251902 (Why is no real title available?)
• scientific article; zbMATH DE number 3385132 (Why is no real title available?)
• M-estimation using penalties or sieves
• Maximum Likelihood Estimation of Misspecified Models
• One-step sparse estimates in nonconcave penalized likelihood models
• Persistene in high-dimensional linear predictor-selection and the virtue of overparametrization
• Probability Inequalities for Sums of Bounded Random Variables
• Robust inference for univariate proportional hazards frailty regression models
• The Adaptive Lasso and Its Oracle Properties
• The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
• Tilting methods for assessing the influence of components in a classifier
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Weak convergence and empirical processes. With applications to statistics

Cited in

(only showing first 100 items - show all)

• Feature screening for time-varying coefficient models with ultrahigh-dimensional longitudinal data
• Factor-Adjusted Regularized Model Selection
• Nonparametric feature screening
• Variable selection in censored quantile regression with high dimensional data
• Conditional feature screening for mean and variance functions in models with multiple-index structure
• Censored cumulative residual independent screening for ultrahigh-dimensional survival data
• Nonparametric independence screening via favored smoothing bandwidth
• Conditional quantile correlation screening procedure for ultrahigh-dimensional varying coefficient models
• Distributed testing and estimation under sparse high dimensional models
• Feature selection of ultrahigh-dimensional covariates with survival outcomes: a selective review
• Ultrahigh dimensional feature screening via projection
• Principal components adjusted variable screening
• Correlation rank screening for ultrahigh-dimensional survival data
• Model free feature screening for ultrahigh dimensional data with responses missing at random
• The use of random-effect models for high-dimensional variable selection problems
• Feature screening for generalized varying coefficient models with application to dichotomous responses
• Adaptive conditional feature screening
• High-dimensional multivariate posterior consistency under global-local shrinkage priors
• A new nonparametric screening method for ultrahigh-dimensional survival data
• Robust feature screening for ultra-high dimensional right censored data via distance correlation
• Fused mean-variance filter for feature screening
• Adjusted Pearson chi-square feature screening for multi-classification with ultrahigh dimensional data
• Model-free conditional independence feature screening for ultrahigh dimensional data
• Model-free feature screening for ultrahigh dimensional censored regression
• Consistency of logistic classifier in abstract Hilbert spaces
• Robust conditional nonparametric independence screening for ultrahigh-dimensional data
• Covariance-insured screening
• Feature screening in ultrahigh-dimensional partially linear models with missing responses at random
• Regression adjustment for treatment effect with multicollinearity in high dimensions
• Modified SCAD penalty for constrained variable selection problems
• Variable screening for ultrahigh dimensional heterogeneous data via conditional quantile correlations
• Model-free feature screening for ultrahigh-dimensional data conditional on some variables
• Variable screening for high dimensional time series
• Hypothesis testing sure independence screening for nonparametric regression
• Detecting rare and faint signals via thresholding maximum likelihood estimators
• Feature screening for nonparametric and semiparametric models with ultrahigh-dimensional covariates
• Nonparametric independence feature screening for ultrahigh-dimensional survival data
• Feature screening for multi-response varying coefficient models with ultrahigh dimensional predictors
• Maximum likelihood estimation in logistic regression models with a diverging number of covariates
• Penalized wavelets: embedding wavelets into semiparametric regression
• Quantile-adaptive model-free variable screening for high-dimensional heterogeneous data
• Model-free feature screening for high-dimensional survival data
• Measuring and testing for interval quantile dependence
• Fused variable screening for massive imbalanced data
• Feature screening for ultrahigh dimensional categorical data with covariates missing at random
• A nonparametric feature screening method for ultrahigh-dimensional missing response
• A note on quantile feature screening via distance correlation
• A distribution-based Lasso for a general single-index model
• SIS
• Feature screening based on distance correlation for ultrahigh-dimensional censored data with covariate measurement error
• An efficient algorithm for joint feature screening in ultrahigh-dimensional Cox's model
• Gini correlation for feature screening
• The de-biased group Lasso estimation for varying coefficient models
• Model-free feature screening via distance correlation for ultrahigh dimensional survival data
• Feature screening for ultrahigh-dimensional survival data when failure indicators are missing at random
• Sparse classification: a scalable discrete optimization perspective
• Conditional screening for ultrahigh-dimensional survival data in case-cohort studies
• Interaction screening via canonical correlation
• High-dimensional variable screening through kernel-based conditional mean dependence
• Asset selection based on high frequency Sharpe ratio
• Surrogate-variable-based model-free feature screening for survival data under the general censoring mechanism
• Interaction identification and clique screening for classification with ultra-high dimensional discrete features
• A general framework for tensor screening through smoothing
• On sufficient variable screening using log odds ratio filter
• Variable screening for varying coefficient models with ultrahigh-dimensional survival data
• Feature screening and FDR control with knockoff features for ultrahigh-dimensional right-censored data
• Independence index sufficient variable screening for categorical responses
• The backbone method for ultra-high dimensional sparse machine learning
• Model-free feature screening for ultrahigh dimensional classification
• Feature screening under missing indicator imputation with non-ignorable missing response
• A procedure of linear discrimination analysis with detected sparsity structure for high-dimensional multi-class classification
• Uniform joint screening for ultra-high dimensional graphical models
• Sequential feature screening for generalized linear models with sparse ultra-high dimensional data
• Robust composite weighted quantile screening for ultrahigh dimensional discriminant analysis
• Ultra-high dimensional variable screening via Gram-Schmidt orthogonalization
• Conditional SIRS for nonparametric and semiparametric models by marginal empirical likelihood
• Dynamic tilted current correlation for high dimensional variable screening
• Sparse regression: scalable algorithms and empirical performance
• Composite quantile regression for ultra-high dimensional semiparametric model averaging
• Forward regression for Cox models with high-dimensional covariates
• Screening and selection for quantile regression using an alternative measure of variable importance
• Sufficient variable selection using independence measures for continuous response
• Feature screening for ultrahigh-dimensional censored data with varying coefficient single-index model
• Feature screening for ultrahigh-dimensional additive logistic models
• Selective inference via marginal screening for high dimensional classification
• Nonparametric screening under conditional strictly convex loss for ultrahigh dimensional sparse data
• Sure independence screening in ultrahigh dimensional generalized additive models
• A two-stage sparse logistic regression for optimal gene selection in high-dimensional microarray data classification
• Adaptively weighted group Lasso for semiparametric quantile regression models
• Joint feature screening for ultra-high-dimensional sparse additive hazards model by the sparsity-restricted pseudo-score estimator
• ARGONAUT: algorithms for global optimization of constrained grey-box computational problems
• Conditional sure independence screening by conditional marginal empirical likelihood
• Bayesian sparse reduced rank multivariate regression
• Sure screening by ranking the canonical correlations
• The cumulative Kolmogorov filter for model-free screening in ultrahigh dimensional data
• Model-free feature screening via a modified composite quantile correlation
• A modified mean-variance feature-screening procedure for ultrahigh-dimensional discriminant analysis
• Nonparametric independence screening for ultra-high dimensional generalized varying coefficient models with longitudinal data
• Consistent tuning parameter selection in high-dimensional group-penalized regression
• High-dimensional influence measure

This page was built for publication: Sure independence screening in generalized linear models with NP-dimensionality

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q140975)