The Kendall interaction filter for variable interaction screening in high dimensional classification problems
From MaRDI portal
Publication:6134401
Abstract: Accounting for important interaction effects can improve prediction of many statistical learning models. Identification of relevant interactions, however, is a challenging issue owing to their ultrahigh-dimensional nature. Interaction screening strategies can alleviate such issues. However, due to heavier tail distribution and complex dependence structure of interaction effects, innovative robust and/or model-free methods for screening interactions are required to better scale analysis of complex and high-throughput data. In this work, we develop a new model-free interaction screening method, termed Kendall Interaction Filter (KIF), for the classification in high-dimensional settings. The KIF method suggests a weighted-sum measure, which compares the overall to the within-cluster Kendall's of pairs of predictors, to select interactive couples of features. The proposed KIF measure captures relevant interactions for the clusters response-variable, handles continuous, categorical or a mixture of continuous-categorical features, and is invariant under monotonic transformations. We show that the KIF measure enjoys the sure screening property in the high-dimensional setting under mild conditions, without imposing sub-exponential moment assumptions on the features' distributions. We illustrate the favorable behavior of the proposed methodology compared to the methods in the same category using simulation studies, and we conduct real data analyses to demonstrate its utility.
Cites work
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- scientific article; zbMATH DE number 3038497 (Why is no real title available?)
- A generic sure independence screening procedure
- Ball Covariance: A Generic Measure of Dependence in Banach Space
- Covariate assisted screening and estimation
- Factor profiled sure independence screening
- Feature screening via distance correlation learning
- Finding predictive gene groups from microarray data
- Grouped variable screening for ultra-high dimensional data for linear model
- Innovated interaction screening for high-dimensional nonlinear classification
- Interaction pursuit in high-dimensional multi-response regression via distance correlation
- Interaction screening for ultrahigh-dimensional data
- Lectures on the Combinatorics of Free Probability
- Measuring and testing dependence by correlation of distances
- Model-free feature screening for ultrahigh dimensional discriminant analysis
- Nonparametric independence screening in sparse ultra-high-dimensional additive models
- On selecting interacting features from high-dimensional data
- Partition-based ultrahigh-dimensional variable screening
- Robust Variable and Interaction Selection for Logistic Regression and General Index Models
- Robust rank correlation based screening
- Sure independence screening for ultrahigh dimensional feature space. With discussion and authors' reply
- Sure independence screening in generalized linear models with NP-dimensionality
- The Kolmogorov filter for variable screening in high-dimensional binary classification
- The fused Kolmogorov filter: a nonparametric model-free screening method
Cited in
(2)
This page was built for publication: The Kendall interaction filter for variable interaction screening in high dimensional classification problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6134401)
