Model-free sure screening via maximum correlation
From MaRDI portal
Abstract: We consider the problem of screening features in an ultrahigh-dimensional setting. Using maximum correlation, we develop a novel procedure called MC-SIS for feature screening, and show that MC-SIS possesses the sure screen property without imposing model or distributional assumptions on the response and predictor variables. Therefore, MC-SIS is a model-free sure independence screening method as in contrast with some other existing model-based sure independence screening methods in the literature. Simulation examples and a real data application are used to demonstrate the performance of MC-SIS as well as to compare MC-SIS with other existing sure screening methods. The results show that MC-SIS outperforms those methods when their model assumptions are violated, and it remains competitive when the model assumptions hold.
Recommendations
- Feature screening via distance correlation learning
- Robust model-free feature screening via quantile correlation
- Model-free feature screening for ultrahigh-dimensional data conditional on some variables
- A generic sure independence screening procedure
- Model-free feature screening via a modified composite quantile correlation
Cites work
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- A Non-Parametric Test of Independence
- A consistent multivariate test of association based on ranks of distances
- Additive regression and other nonparametric models
- An extremal property of rectangular distributions
- Detecting novel associations in large data sets
- Discussion of: Brownian distance covariance
- Estimating Optimal Transformations for Multiple Regression and Correlation
- Feature screening via distance correlation learning
- Local asymptotics for regression splines and confidence regions
- Measuring and testing dependence by correlation of distances
- Nonparametric Independence Screening in Sparse Ultra-High-Dimensional Varying Coefficient Models
- Nonparametric independence screening in sparse ultra-high-dimensional additive models
- On measures of dependence
- On the Maximum Correlation Coefficient
- On the maximal correlation coefficient
- Perturbation theory for linear operators.
- Rates of convergence for spline estimates of additive principal components
- Rates of convergence for the estimates of the optimal transformations of variables
- Regularization and Variable Selection Via the Elastic Net
- Remarks on the maximum correlation coefficient
- Sure independence screening for ultrahigh dimensional feature space. With discussion and authors' reply
- Sure independence screening in generalized linear models with NP-dimensionality
- Ultrahigh dimensional feature selection: beyond the linear model
- Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
- Variable selection in nonparametric additive models
- Weak convergence and empirical processes. With applications to statistics
Cited in
(22)- Feature screening via distance correlation learning
- An iterative approach to distance correlation-based sure independence screening
- Four simple axioms of dependence measures
- A consistent variable screening procedure with family-wise error control
- Variable screening for high dimensional time series
- Feature screening for network autoregression model
- Model-free feature screening for ultrahigh dimensional data through a modified Blum-Kiefer-Rosenblatt correlation
- A generic sure independence screening procedure
- Robust dependence measure for detecting associations in large data set
- Model-free feature screening via a modified composite quantile correlation
- Lancaster correlation: a new dependence measure linked to maximum correlation
- Model‐free conditional screening for ultrahigh‐dimensional survival data via conditional distance correlation
- Grouped feature screening for ultrahigh-dimensional classification via Gini distance correlation
- Sure screening by ranking the canonical correlations
- Robust model-free feature screening via quantile correlation
- Sure explained variability and independence screening
- Model-free conditional screening via conditional distance correlation
- Model-Free Feature Screening and FDR Control With Knockoff Features
- A New Model-Free Feature Screening Procedure for Ultrahigh-Dimensional Interval-Censored Failure Time Data
- Distribution-free and model-free multivariate feature screening via multivariate rank distance correlation
- Consistent Screening Procedures in High-dimensional Binary Classification
- Linear screening for high-dimensional computer experiments
This page was built for publication: Model-free sure screening via maximum correlation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q276978)
