Bibliographic Details
| Title: |
Sparse quantile regression via ℓ0-penalty. |
| Authors: |
Honda, Toshio1 (AUTHOR) t.honda@r.hit-u.ac.jp, Wu, Wei-Ying2 (AUTHOR) wuweiying1011@gmail.com |
| Source: |
Annals of the Institute of Statistical Mathematics. Feb2026, Vol. 78 Issue 1, p141-173. 33p. |
| Subject Terms: |
*Quantile regression, *High-dimensional model representation, *Linear statistical models, *Tikhonov regularization, *Statistical models |
| Abstract: |
We consider model selection via ℓ 0 -penalty for high-dimensional sparse quantile regression models. This procedure is almost equivalent to model selection via information criterion due to similarity in penalty. We deal with linear models, additive models, and varying coefficient models in a unified way and establish the model selection consistency results rigorously when the size of the relevant index set goes to infinity. The treatment of this situation is challenging and the theoretical novelty of our results is important because such information criteria are commonly used. We consider two different setups and propose tuning parameters in the ℓ 0 -penalty. Besides, we propose a feasible algorithm for computation of our estimator and the numerical study results are presented. [ABSTRACT FROM AUTHOR] |
| Database: |
Academic Search Index |