Feature screening plays an important role in ultrahigh dimensional data analysis.This paper is concerned with conditional feature screening when one is interested in detecting the association between the response and ...Feature screening plays an important role in ultrahigh dimensional data analysis.This paper is concerned with conditional feature screening when one is interested in detecting the association between the response and ultrahigh dimensional predictors(e.g.,genetic makers)given a low-dimensional exposure variable(such as clinical variables or environmental variables).To this end,we first propose a new index to measure conditional independence,and further develop a conditional screening procedure based on the newly proposed index.We systematically study the theoretical property of the proposed procedure and establish the sure screening and ranking consistency properties under some very mild conditions.The newly proposed screening procedure enjoys some appealing properties.(a)It is model-free in that its implementation does not require a specification on the model structure;(b)it is robust to heavy-tailed distributions or outliers in both directions of response and predictors;and(c)it can deal with both feature screening and the conditional screening in a unified way.We study the finite sample performance of the proposed procedure by Monte Carlo simulations and further illustrate the proposed method through two real data examples.展开更多
基金supported by National Science Foundation of USA (Grant No. P50 DA039838)the Program of China Scholarships Council (Grant No. 201506040130)+6 种基金 National Natural Science Foundation of China (Grant No. 11401497)the Scientific Research Foundation for the Returned Overseas Chinese ScholarsState Education Ministry, the National Key Basic Research Development Program of China (Grant No. 2010CB950703)the Fundamental Research Funds for the Central UniversitiesNational Institute on Drug AbuseNational Institutes of Health (Grants Nos. P50 DA036107 and P50 DA039838)National Science Foundation of USA (Grant No. DMS 1512422)
文摘Feature screening plays an important role in ultrahigh dimensional data analysis.This paper is concerned with conditional feature screening when one is interested in detecting the association between the response and ultrahigh dimensional predictors(e.g.,genetic makers)given a low-dimensional exposure variable(such as clinical variables or environmental variables).To this end,we first propose a new index to measure conditional independence,and further develop a conditional screening procedure based on the newly proposed index.We systematically study the theoretical property of the proposed procedure and establish the sure screening and ranking consistency properties under some very mild conditions.The newly proposed screening procedure enjoys some appealing properties.(a)It is model-free in that its implementation does not require a specification on the model structure;(b)it is robust to heavy-tailed distributions or outliers in both directions of response and predictors;and(c)it can deal with both feature screening and the conditional screening in a unified way.We study the finite sample performance of the proposed procedure by Monte Carlo simulations and further illustrate the proposed method through two real data examples.