Robust and consistent variable selection in high-dimensional generalized linear models

被引：21

作者：

Avella-Medina, Marco ^{[1
]}

Ronchetti, Elvezio ^{[2
]}

机构：

[1] MIT, Sloan Sch Management, 30 Mem Dr, Cambridge, MA 02142 USA

[2] Univ Geneva, Res Ctr Stat, Blvd Pont Arve 40, CH-1205 Geneva, Switzerland

来源：

BIOMETRIKA | 2018年 / 105卷 / 01期

基金：

瑞士国家科学基金会;

关键词：

Contamination neighbourhood; Generalized linear model; Infinitesimal robustness; Lasso; Oracle estimator; Robust quasilikelihood; NONCONCAVE PENALIZED LIKELIHOOD; REGRESSION SHRINKAGE; CONFIDENCE-INTERVALS; ADAPTIVE LASSO; INFERENCE; ESTIMATORS; REGULARIZATION;

D O I：

10.1093/biomet/asx070

中图分类号：

Q [生物科学];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

Generalized linear models are popular for modelling a large variety of data. We consider variable selection through penalized methods by focusing on resistance issues in the presence of outlying data and other deviations from assumptions. We highlight the weaknesses of widely-used penalized M-estimators, propose a robust penalized quasilikelihood estimator, and show that it enjoys oracle properties in high dimensions and is stable in a neighbourhood of the model. We illustrate its finite-sample performance on simulated and real data.

引用

页码：31 / 44

页数：14

共 50 条

[21] Consistent significance controlled variable selection in high-dimensional regression
Zambom, Adriano Zanin
Kim, Jongwook
[J]. STAT, 2018, 7 (01):
[22] Efficient test-based variable selection for high-dimensional linear models
Gong, Siliang
Zhang, Kai
Liu, Yufeng
[J]. JOURNAL OF MULTIVARIATE ANALYSIS, 2018, 166 : 17 - 31
[23] Variable selection in multivariate linear models with high-dimensional covariance matrix estimation
Perrot-Dockes, Marie
Levy-Leduc, Celine
Sansonnet, Laure
Chiquet, Julien
[J]. JOURNAL OF MULTIVARIATE ANALYSIS, 2018, 166 : 78 - 97
[24] A STEPWISE REGRESSION METHOD AND CONSISTENT MODEL SELECTION FOR HIGH-DIMENSIONAL SPARSE LINEAR MODELS
Ing, Ching-Kang
Lai, Tze Leung
[J]. STATISTICA SINICA, 2011, 21 (04) : 1473 - 1513
[25] Variable selection for high-dimensional generalized linear model with block-missing data
He, Yifan
Feng, Yang
Song, Xinyuan
[J]. SCANDINAVIAN JOURNAL OF STATISTICS, 2023, 50 (03) : 1279 - 1297
[26] Variable selection in the high-dimensional continuous generalized linear model with current status data
Tian, Guo-Liang
Wang, Mingqiu
Song, Lixin
[J]. JOURNAL OF APPLIED STATISTICS, 2014, 41 (03) : 467 - 483
[27] The robust desparsified lasso and the focused information criterion for high-dimensional generalized linear models
Pandhare, S. C.
Ramanathan, T. V.
[J]. STATISTICS, 2023, 57 (01) : 1 - 25
[28] GREEDY VARIABLE SELECTION FOR HIGH-DIMENSIONAL COX MODELS
Lin, Chien-Tong
Cheng, Yu-Jen
Ing, Ching-Kang
[J]. STATISTICA SINICA, 2023, 33 : 1697 - 1719
[29] High-dimensional robust inference for censored linear models
Jiayu Huang
Yuanshan Wu
[J]. Science China Mathematics, 2024, 67 : 891 - 918
[30] Learning High-Dimensional Generalized Linear Autoregressive Models
Hall, Eric C.
Raskutti, Garvesh
Willett, Rebecca M.
[J]. IEEE TRANSACTIONS ON INFORMATION THEORY, 2019, 65 (04) : 2401 - 2422

← 1 2 3 4 5 →