Sure independence screening in generalized linear models with NP-dimensionality

DOI10.1214/10-AOS798MaRDI QIDQ140975zbMATHOpenAlexFDO

Authors Jianqing Fan, Rui Song, 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

variable selectiongeneralized linear modelsindependent learningsure independent screening

Mathematics Subject Classification ID

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

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

• 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?)
• 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
• 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)

• Interaction screening for ultrahigh-dimensional data
• Sure independence screening in ultrahigh dimensional generalized additive models
• The fused Kolmogorov filter: a nonparametric model-free screening method
• Quantile-adaptive model-free variable screening for high-dimensional heterogeneous data
• Bayesian high-dimensional screening via MCMC
• Marginal empirical likelihood and sure independence feature screening
• Martingale difference correlation and its use in high-dimensional variable screening
• Principled sure independence screening for Cox models with ultra-high-dimensional covariates
• Adaptive Lasso estimators for ultrahigh dimensional generalized linear models
• Robust feature screening procedures for single and mixed types of data
• A distribution-based Lasso for a general single-index model
• Robust sure independence screening for ultrahigh dimensional non-normal data
• Model-free sure screening via maximum correlation
• Variable selection in censored quantile regression with high dimensional data
• Goodness-of-fit testing-based selection for large-\(p\)-small-\(n\) problems: a two-stage ranking approach
• A sure independence screening procedure for ultra-high dimensional partially linear additive models
• Sure independence screening for ultrahigh dimensional feature space. With discussion and authors' reply
• Local independence feature screening for nonparametric and semiparametric models by marginal empirical likelihood
• Sparse regression: scalable algorithms and empirical performance
• High-dimensional influence measure
• Robust rank correlation based screening
• Covariate assisted screening and estimation
• A regularization-based adaptive test for high-dimensional GLMs
• Ranking-based variable selection for high-dimensional data
• Feature selection when there are many influential features
• Ultrahigh dimensional variable selection through the penalized maximum trimmed likelihood estimator
• Linear hypothesis testing in dense high-dimensional linear models
• Testing covariates in high dimension linear regression with latent factors
• Feature screening for time-varying coefficient models with ultrahigh-dimensional longitudinal data
• Ultrahigh dimensional feature selection: beyond the linear model
• Greedy forward regression for variable screening
• The de-biased group Lasso estimation for varying coefficient models
• Estimating false discovery proportion under arbitrary covariance dependence
• Feature screening via distance correlation learning
• Model-free conditional independence feature screening for ultrahigh dimensional data
• Model-free feature screening for ultrahigh dimensional censored regression
• Feature selection for varying coefficient models with ultrahigh-dimensional covariates
• Factor-Adjusted Regularized Model Selection
• Going beyond oracle property: selection consistency and uniqueness of local solution of the generalized linear model
• Feature screening for generalized varying coefficient models with application to dichotomous responses
• Dummy endogenous treatment effect estimation using high‐dimensional instrumental variables
• SIS
• Model selection and structure specification in ultra-high dimensional generalised semi-varying coefficient models
• Consistent Screening Procedures in High-dimensional Binary Classification
• ARGONAUT: algorithms for global optimization of constrained grey-box computational problems
• Maximum likelihood estimation in logistic regression models with a diverging number of covariates
• Screening-based Bregman divergence estimation with NP-dimensionality
• Measuring and testing for interval quantile dependence
• Analyzing large datasets with bootstrap penalization
• Penalized wavelets: embedding wavelets into semiparametric regression
• Likelihood ratio test in multivariate linear regression: from low to high dimension
• scientific article; zbMATH DE number 7750673 (Why is no real title available?)
• Robust rank screening for ultrahigh dimensional discriminant analysis
• Variable selection for additive partial linear quantile regression with missing covariates
• Robust sure independence screening for nonpolynomial dimensional generalized linear models
• Nonparametric independence screening in sparse ultra-high-dimensional additive models
• The use of random-effect models for high-dimensional variable selection problems
• Detecting rare and faint signals via thresholding maximum likelihood estimators
• High-dimensional Bayesian inference in nonparametric additive models
• Robust feature screening for varying coefficient models via quantile partial correlation
• Independent feature screening for ultrahigh-dimensional models with interactions
• Model-free feature screening for ultrahigh-dimensional data conditional on some variables
• Selective inference via marginal screening for high dimensional classification
• Profile forward regression screening for ultra-high dimensional semiparametric varying coefficient partially linear models
• Nonparametric Independence Screening in Sparse Ultra-High-Dimensional Varying Coefficient Models
• High-dimensional multivariate posterior consistency under global-local shrinkage priors
• Nonparametric independence screening for ultra-high dimensional generalized varying coefficient models with longitudinal data
• Ultrahigh dimensional time course feature selection
• Consistent tuning parameter selection in high-dimensional group-penalized regression
• Independent screening in high-dimensional exponential family predictors’ space
• Sparse classification: a scalable discrete optimization perspective
• Adaptively weighted group Lasso for semiparametric quantile regression models
• Bayesian Subset Modeling for High-Dimensional Generalized Linear Models
• Nonparametric feature screening
• Interaction screening via canonical correlation
• Surrogate-variable-based model-free feature screening for survival data under the general censoring mechanism
• Feature screening for ultrahigh-dimensional survival data when failure indicators are missing at random
• Modified SCAD penalty for constrained variable selection problems
• Feature screening and FDR control with knockoff features for ultrahigh-dimensional right-censored data
• Independence index sufficient variable screening for categorical responses
• Copula-based Partial Correlation Screening: a Joint and Robust Approach
• The backbone method for ultra-high dimensional sparse machine learning
• Conditional characteristic feature screening for massive imbalanced data
• L2RM: Low-Rank Linear Regression Models for High-Dimensional Matrix Responses
• Feature screening with large‐scale and high‐dimensional survival data
• Asset selection based on high frequency Sharpe ratio
• Ultrahigh dimensional feature screening for additive model with multivariate response
• A procedure of linear discrimination analysis with detected sparsity structure for high-dimensional multi-class classification
• A two-stage sparse logistic regression for optimal gene selection in high-dimensional microarray data classification
• An efficient algorithm for joint feature screening in ultrahigh-dimensional Cox's model
• Feature screening based on distance correlation for ultrahigh-dimensional censored data with covariate measurement error
• Robust composite weighted quantile screening for ultrahigh dimensional discriminant analysis
• Conditional SIRS for nonparametric and semiparametric models by marginal empirical likelihood
• Ultra-high dimensional variable screening via Gram-Schmidt orthogonalization
• An iterative approach to distance correlation-based sure independence screening
• Robust feature screening for high-dimensional survival data
• A general framework for tensor screening through smoothing
• Uniform joint screening for ultra-high dimensional graphical models
• On sufficient variable screening using log odds ratio filter
• Feature screening for network autoregression model

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)