Journal article
GSDAR: a fast Newton algorithm for l(0) regularized generalized linearmodels with statistical guarantee
Computational statistics, Vol.37(1), pp.507-533
03/29/2021
DOI: 10.1007/s00180-021-01098-z
Abstract
We propose a fast Newton algorithm for l(0) regularized high-dimensional generalized linear models based on support detection and root finding. We refer to the proposed method as GSDAR. GSDAR is developed based on the KKT conditions for l(0)-penalized maximum likelihood estimators and generates a sequence of solutions of the KKT system iteratively. We show that GSDAR can be equivalently formulated as a generalized Newton algorithm. Under a restricted invertibility condition on the likelihood function and a sparsity condition on the regression coefficient, we establish an explicit upper bound on the estimation errors of the solution sequence generated by GSDAR in supremum norm and show that it achieves the optimal order in finite iterations with high probability. Moreover, we show that the oracle estimator can be recovered with high probability if the target signal is above the detectable level. These results directly concern the solution sequence generated from the GSDAR algorithm, instead of a theoretically defined global solution. We conduct simulations and real data analysis to illustrate the effectiveness of the proposed method.
Details
- Title: Subtitle
- GSDAR: a fast Newton algorithm for l(0) regularized generalized linearmodels with statistical guarantee
- Creators
- Jian Huang - University of IowaYuling Jiao - Wuhan UniversityLican Kang - Wuhan UniversityJin Liu - National University of SingaporeYanyan Liu - Wuhan UniversityXiliang Lu - Wuhan University
- Resource Type
- Journal article
- Publication Details
- Computational statistics, Vol.37(1), pp.507-533
- Publisher
- Springer Nature
- DOI
- 10.1007/s00180-021-01098-z
- ISSN
- 0943-4062
- eISSN
- 1613-9658
- Number of pages
- 27
- Grant note
- 11871474; 11971362; 11471253; 91630313 / National Science Foundation of China; National Natural Science Foundation of China (NSFC) WBS: R-913-200-098-263; MOE2016-T2-2-029 / Duke-NUS Graduate Medical School from Ministry of Eduction, Singapore DMS-1916199 / U.S. National Science Foundation; National Science Foundation (NSF) KLATASDSMOE of China
- Language
- English
- Date published
- 03/29/2021
- Academic Unit
- Statistics and Actuarial Science
- Record Identifier
- 9984257721402771
Metrics
15 Record Views