Local independence feature screening for nonparametric and semiparametric models by marginal empirical likelihood
arXiv:1502.07061 · doi:10.1214/15-AOS1374
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.
arXiv commentPublished at http://dx.doi.org/10.1214/15-AOS1374 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
References in corpus (7)
- Sure independence screening in generalized linear models with NP-dimensionality
- Component selection and smoothing in multivariate nonparametric regression
- Variable selection in nonparametric additive models
- Robust rank correlation based screening
- Extending the scope of empirical likelihood
- Marginal empirical likelihood and sure independence feature screening
- High dimensional generalized empirical likelihood for moment restrictions with dependent data
Cited by in corpus (5)
- Simulation-Based Hypothesis Testing of High Dimensional Means Under Covariance Heterogeneity
- A new scope of penalized empirical likelihood with high-dimensional estimating equations
- Cramér-type moderate deviations for Studentized two-sample -statistics with applications
- High-dimensional empirical likelihood inference
- Nonparametric Independence Screening via Favored Smoothing Bandwidth