Covariate Information Number for Feature Screening in Ultrahigh-Dimensional Supervised Problems.
Covariate Information Number for Feature Screening in Ultrahigh-Dimensional Supervised Problems.
复制标题
DOI:
10.1080/01621459.2020.1864380
复制
发表时间:
2022
影响因子:
3.7
通讯作者:
Li, Runze
中科院分区:
文献类型:
--
作者:
Nandy, Debmalya;Chiaromonte, Francesca;Li, Runze
关键词:
Contemporary high-throughput experimental and surveying techniques give rise to ultrahigh-dimensional supervised problems with sparse signals; that is, a limited number of observations (n), each with a very large number of covariates (p >> n), only a small share of which is truly associated with the response. In these settings, major concerns on computational burden, algorithmic stability, and statistical accuracy call for substantially reducing the feature space by eliminating redundant covariates before the use of any sophisticated statistical analysis. Along the lines of Sure Independence Screening and other model- and correlation-based feature screening methods, we propose a model-free procedure called Covariate Information Number - Sure Independence Screening (CIS). CIS uses a marginal utility connected to the notion of the traditional Fisher Information, possesses the sure screening property, and is applicable to any type of response (features) with continuous features (response). Simulations and an application to transcriptomic data on rats reveal the comparative strengths of CIS over some popular feature screening methods.
登录
查看更多内容
DOI:
10.1080/01621459.2013.850086
发表时间:
2014-01-01
影响因子:
3.7
作者:
Liu J;Li R;Wu R
通讯作者:
Wu R
影响因子:
4.5
作者:
Huang J;Horowitz JL;Wei F
通讯作者:
Wei F
影响因子:
3.7
作者:
Fan J;Feng Y;Song R
通讯作者:
Song R
DOI:
10.1073/pnas.0600158103
发表时间:
2006-04-18
影响因子:
11.1
作者:
Chiang, AP;Beck, JS;Sheffield, VC
通讯作者:
Sheffield, VC
DOI:
10.1080/07350015.2013.863158
发表时间:
2014
期刊:
Journal of business & economic statistics : a publication of the American Statistical Association
影响因子:
--
作者:
Huang D;Li R;Wang H
通讯作者:
Wang H