A global two-stage algorithm for non-convex penalized high-dimensional linear regression problems

被引:1
|
作者
Li, Peili [1 ]
Liu, Min [2 ]
Yu, Zhou [1 ]
机构
[1] East China Normal Univ, KLATASDS MOE, Sch Stat, Shanghai 200062, Peoples R China
[2] Wuhan Univ, Sch Math & Stat, Wuhan 430072, Peoples R China
基金
中国国家自然科学基金; 国家重点研发计划;
关键词
High-dimensional linear regression; Global convergence; Two-stage algorithm; Primal dual active set with continuation algorithm; Difference of convex functions; COORDINATE DESCENT ALGORITHMS; ACTIVE SET ALGORITHM; VARIABLE SELECTION; LIKELIHOOD;
D O I
10.1007/s00180-022-01249-w
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
By the asymptotic oracle property, non-convex penalties represented by minimax concave penalty (MCP) and smoothly clipped absolute deviation (SCAD) have attracted much attentions in high-dimensional data analysis, and have been widely used in signal processing, image restoration, matrix estimation, etc. However, in view of their non-convex and non-smooth characteristics, they are computationally challenging. Almost all existing algorithms converge locally, and the proper selection of initial values is crucial. Therefore, in actual operation, they often combine a warm-starting technique to meet the rigid requirement that the initial value must be sufficiently close to the optimal solution of the corresponding problem. In this paper, based on the DC (difference of convex functions) property of MCP and SCAD penalties, we aim to design a global two-stage algorithm for the high-dimensional least squares linear regression problems. A key idea for making the proposed algorithm to be efficient is to use the primal dual active set with continuation (PDASC) method to solve the corresponding sub-problems. Theoretically, we not only prove the global convergence of the proposed algorithm, but also verify that the generated iterative sequence converges to a d-stationary point. In terms of computational performance, the abundant research of simulation and real data show that the algorithm in this paper is superior to the latest SSN method and the classic coordinate descent (CD) algorithm for solving non-convex penalized high-dimensional linear regression problems.
引用
收藏
页码:871 / 898
页数:28
相关论文
共 50 条
  • [21] An Iterative Coordinate Descent Algorithm for High-Dimensional Nonconvex Penalized Quantile Regression
    Peng, Bo
    Wang, Lan
    JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2015, 24 (03) : 676 - 694
  • [22] Analytical Study of Momentum-Based Acceleration Methods in Paradigmatic High-Dimensional Non-Convex Problems
    Mannelli, Stefano Sarao
    Urbani, Pierfrancesco
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34
  • [23] CONVERGENCE OF A GREEDY ALGORITHM FOR HIGH-DIMENSIONAL CONVEX NONLINEAR PROBLEMS
    Cances, Eric
    Ehrlacher, Virginie
    Lelievre, Tony
    MATHEMATICAL MODELS & METHODS IN APPLIED SCIENCES, 2011, 21 (12): : 2433 - 2467
  • [24] High-dimensional local linear regression under sparsity and convex losses
    Cheung, Kin Yap
    Lee, Stephen M. S.
    ELECTRONIC JOURNAL OF STATISTICS, 2024, 18 (01): : 803 - 847
  • [25] Two-Stage Data-Driven Evolutionary Optimization for High-Dimensional Expensive Problems
    Zhen, Huixiang
    Gong, Wenyin
    Wang, Ling
    Ming, Fei
    Liao, Zuowen
    IEEE TRANSACTIONS ON CYBERNETICS, 2023, 53 (04) : 2368 - 2379
  • [26] Penalized Ordinal Regression Methods for Predicting Stage of Cancer in High-Dimensional Covariate Spaces
    Gentry, Amanda Elswick
    Jackson-Cook, Colleen K.
    Lyon, Debra E.
    Archer, Kellie J.
    CANCER INFORMATICS, 2015, 14 : 201 - 208
  • [27] A two-stage clonal selection algorithm for local feature selection on high-dimensional data
    Wang, Yi
    Tian, Hao
    Li, Tao
    Liu, Xiaojie
    INFORMATION SCIENCES, 2024, 677
  • [28] Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
    Dauphin, Yann N.
    Pascanu, Razvan
    Gulcehre, Caglar
    Cho, Kyunghyun
    Ganguli, Surya
    Bengio, Yoshua
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 27 (NIPS 2014), 2014, 27
  • [29] Assessment of a non-adaptive deterministic global optimization algorithm for problems with low-dimensional non-convex subspaces
    Kearfott, Ralph Baker
    Castille, Jessie M.
    Tyagi, Gaurav
    OPTIMIZATION METHODS & SOFTWARE, 2014, 29 (02): : 430 - 441
  • [30] A two-stage sequential conditional selection approach to sparse high-dimensional multivariate regression models
    Zehua Chen
    Yiwei Jiang
    Annals of the Institute of Statistical Mathematics, 2020, 72 : 65 - 90