Conditional variable screening via ordinary least squares projection
Preprint
Abstract
To deal with the growing challenge from high dimensional data, we propose a conditional variable screening method for linear models named as conditional screening via ordinary least squares projection (COLP). COLP can take advantage of prior knowledge concerning certain active predictors by eliminating the adverse impacts from their coefficients in the estimation of remaining ones and thus significantly enhance the screening accuracy. We then prove the sure screening property of COLP under reasonable assumptions. Moreover, based on the conditional approach, we introduce an iterative algorithm named as forward screening via ordinary least squares projection (FOLP), which not only could utilize the prior knowledge more effectively, but also has promising performances when no prior information is available applying a data-driven conditioning set. The competence of COLP and FOLP is fully demonstrated in extensive numerical studies and the application to a leukemia dataset.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- Conditional Sure Independence ScreeningJournal of the American Statistical Association, 2016
- Feature Screening via Distance Correlation LearningJournal of the American Statistical Association, 2012
- Introduction to the non-asymptotic analysis of random matrices2012
- Projection Matrices, Generalized Inverse Matrices, and Singular Value Decomposition2011
- Sure Independence Screening for Ultrahigh Dimensional Feature SpaceJournal of the Royal Statistical Society Series B: Statistical Methodology, 2008
- Extended Bayesian information criteria for model selection with large model spacesBiometrika, 2008
- Regularization and Variable Selection Via the Elastic NetJournal of the Royal Statistical Society Series B: Statistical Methodology, 2005
- The percentage of myeloperoxidase-positive blast cells is a strong independent prognostic factor in acute myeloid leukemia, even in the patients with normal karyotypeLeukemia, 2003
- Diagnosis of multiple cancer types by shrunken centroids of gene expressionProceedings of the National Academy of Sciences, 2002
- Regression Shrinkage and Selection Via the LassoJournal of the Royal Statistical Society Series B: Statistical Methodology, 1996