A significance test for graph-constrained estimation
From MaRDI portal
(Redirected from Publication:5739274)
A significance test for graph-constrained estimation (scientific article; zbMATH DE number 6603738)
A significance test for graph-constrained estimation (scientific article; zbMATH DE number 6603738)
Abstract: Graph-constrained estimation methods encourage similarities among neighboring covariates presented as nodes on a graph, which can result in more accurate estimations, especially in high dimensional settings. Variable selection approaches can then be utilized to select a subset of variables that are associated with the response. However, existing procedures do not provide measures of uncertainty of the estimates. Moreover, the vast majority of existing approaches assume that available graphs accurately capture the association among covariates; violating this assumption could severely hurt the reliability of the resulting estimates. In this paper, we present an inference framework, called the Grace test, which simultaneously produces coefficient estimates and corresponding -values while incorporating the external graph information. We show, both theoretically and via numerical studies, that the proposed method asymptotically controls the type-I error rate regardless of the choice of the graph. When the underlying graph is informative, the Grace test is asymptotically more powerful than similar tests that ignore external information. We further propose a more general Grace-ridge test that results in a higher power than the Grace test when the choice of the graph is not fully informative. Our numerical studies show that as long as the graph is reasonably informative, the proposed testing methods deliver improved statistical power over existing inference procedures that ignore external information.
Recommendations
- Variable selection and regression analysis for graph-structured covariates with an application to genomics
- Graph structured sparse subset selection
- Sparse estimation of conditional graphical models with application to gene networks
- Graph estimation with joint additive models
- Sparse Laplacian shrinkage with the graphical Lasso estimator for regression problems
Cites work
- scientific article; zbMATH DE number 3624650 (Why is no real title available?)
- scientific article; zbMATH DE number 1347881 (Why is no real title available?)
- Estimation in high-dimensional linear models with deterministic design matrices
- Feature selection guided by structural information
- Incorporating Predictor Network in Penalized Regression with Application to Microarray Data
- Network enrichment analysis in complex experiments
- On asymptotically optimal confidence regions and tests for high-dimensional models
- Scaled sparse linear regression
- Semiparametric Regression of Multidimensional Genetic Pathway Data: Least‐Squares Kernel Machines and Linear Mixed Models
- Simultaneous analysis of Lasso and Dantzig selector
- Simultaneous supervised clustering and feature selection over a graph
- Statistical significance in high-dimensional linear models
- Structured penalties for functional linear models -- partially empirical eigenvectors for regression
- The control of the false discovery rate in multiple testing under dependency.
- The sparse Laplacian shrinkage estimator for high-dimensional regression
- Variable selection and regression analysis for graph-structured covariates with an application to genomics
Cited in
(14)- On the Use of Minimum Penalties in Statistical Learning
- Generalized matrix decomposition regression: estimation and inference for two-way structured data
- Power of edge exclusion tests in graphical Gaussian models
- Kernel-penalized regression for analysis of microbiome data
- Joint estimation of precision matrices in heterogeneous populations
- Detecting clusters in multivariate response regression
- Connectivity‐informed adaptive regularization for generalized outcomes
- High-dimensional Gaussian graphical models on network-linked data
- Gene-gene interaction analysis incorporating network information via a structured Bayesian approach
- Doubly regularized generalized linear models for spatial observations with high-dimensional covariates
- Graph-based sparse linear discriminant analysis for high-dimensional classification
- In defense of the indefensible: a very naïve approach to high-dimensional inference
- Network differential connectivity analysis
- A cluster elastic net for multivariate regression
This page was built for publication: A significance test for graph-constrained estimation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5739274)
