Local independence feature screening for nonparametric and semiparametric models by marginal empirical likelihood
Authors:
- Jinyuan ChangJCJinyuan Chang
- Cheng Yong TangCTCheng Yong Tang
- Yichao WuYWYichao Wu
Preprint
Abstract
We consider an independence feature screening technique for identifying explanatory variables that locally contribute to the response variable in high-dimensional regression analysis. Without requiring a specific parametric form of the underlying data model, our approach accommodates a wide spectrum of nonparametric and semiparametric model families. To detect the local contributions of explanatory variables, our approach constructs empirical likelihood locally in conjunction with marginal nonparametric regressions. Since our approach actually requires no estimation, it is advantageous in scenarios such as the single-index models where even specification and identification of a marginal model is an issue. By automatically incorporating the level of variation of the nonparametric regression and directly assessing the strength of data evidence supporting local contribution from each explanatory variable, our approach provides a unique perspective for solving feature screening problems. Theoretical analysis shows that our approach can handle data dimensionality growing exponentially with the sample size. With extensive theoretical illustrations and numerical examples, we show that the local independence screening approach performs promisingly.Keywords
All Related Versions
This publication has 27 references indexed in Scilit:
- High dimensional generalized empirical likelihood for moment restrictions with dependent dataJournal of Econometrics, 2015
- Nonparametric Independence Screening in Sparse Ultra-High-Dimensional Varying Coefficient ModelsJournal of the American Statistical Association, 2014
- Feature Selection for Varying Coefficient Models With Ultrahigh-Dimensional CovariatesJournal of the American Statistical Association, 2014
- Feature Screening via Distance Correlation LearningJournal of the American Statistical Association, 2012
- A Generalized Linear Model With “Gaussian” Regressor Variables2011
- A review on empirical likelihood methods for regressionTEST, 2009
- Effects of data dimension on empirical likelihoodBiometrika, 2009
- Varying-coefficient models and basis function approximations for the analysis of repeated measurementsBiometrika, 2002
- Empirical likelihood confidence intervals for local linear smoothersBiometrika, 2000
- Empirical likelihood confidence intervals for nonparametric density estimationBiometrika, 1996