TY - JOUR
T1 - Joint Model-free Feature Screening for Ultra-High Dimensional Semi-Competing Risks Data
AU - Lu, Shuiyun
AU - Chen, Xiaolin
AU - Xu, Sheng
AU - Liu, Catherine Chunling
N1 - Funding Information:
This work was conducted while Xiaolin Chen was visiting Department of Statistics, Rutgers University. Xiaolin Chen wants to express his gratitude to Prof. Min-ge Xie for his hospitality during Xiaolin’s visit at Rutgers University. The authors are grateful to the co-editor, anonymous associate editor and two anonymous reviewers for their helpful and insightful comments. Xiaolin Chen’s research is supported by the National Natural Science Foundation of China ( 11771250 and 11501573 ) and a grant from China Scholarship Council . Sheng Xu’s research is supported by The Hong Kong Polytechnic University Ph.D. Studentship. Chunling Liu’s research is supported by the General Research Fund of Hong Kong ( 15327216 ).
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2020/7
Y1 - 2020/7
N2 - High-dimensional semi-competing risks data consisting of two probably correlated events, namely terminal event and non-terminal event, arise commonly in many biomedical studies. However, the corresponding statistical analysis is rarely investigated. A joint model-free feature screening procedure for both terminal and non-terminal events is proposed, which could allow the associated covariates to be in an ultra-high dimensional feature space. The joint screening utility is constructed from distance correlation between each predictor’s survival function and joint survival function of terminal and non-terminal events. Under rather mild technical assumptions, it is demonstrated that the proposed joint feature screening procedure enjoys sure screening and consistency in ranking properties. An adaptive threshold rule is further suggested to simultaneously identify important covariates and determine number of these covariates. Extensive numerical studies are conducted to examine the finite-sample performance of the proposed methods. Lastly, the suggested joint feature screening procedure is illustrated through a real example.
AB - High-dimensional semi-competing risks data consisting of two probably correlated events, namely terminal event and non-terminal event, arise commonly in many biomedical studies. However, the corresponding statistical analysis is rarely investigated. A joint model-free feature screening procedure for both terminal and non-terminal events is proposed, which could allow the associated covariates to be in an ultra-high dimensional feature space. The joint screening utility is constructed from distance correlation between each predictor’s survival function and joint survival function of terminal and non-terminal events. Under rather mild technical assumptions, it is demonstrated that the proposed joint feature screening procedure enjoys sure screening and consistency in ranking properties. An adaptive threshold rule is further suggested to simultaneously identify important covariates and determine number of these covariates. Extensive numerical studies are conducted to examine the finite-sample performance of the proposed methods. Lastly, the suggested joint feature screening procedure is illustrated through a real example.
KW - Clayton copula
KW - Distance correlation
KW - Feature screening
KW - Semi-competing risks data
KW - Ultra-high dimensionality
UR - http://www.scopus.com/inward/record.url?scp=85081019100&partnerID=8YFLogxK
U2 - 10.1016/j.csda.2020.106942
DO - 10.1016/j.csda.2020.106942
M3 - Journal article
SN - 0167-9473
VL - 147
SP - 1
EP - 20
JO - Computational Statistics and Data Analysis
JF - Computational Statistics and Data Analysis
M1 - 106942
ER -