Square-Root LASSO for High-Dimensional Sparse Linear Systems with Weakly Dependent Errors

被引：5
作者：
Xie, Fang [1 ,2 ]
Xiao, Zhijie [3 ,4 ]
机构：
[1] Univ Macau, Fac Sci & Technol, Dept Math, Taipa, Macao, Peoples R China
[2] UM Zhuhai Res Inst, Zhuhai, Peoples R China
[3] Boston Coll, Dept Econ, 140 Commonwealth Ave, Chestnut Hill, MA 02467 USA
[4] Shandong Univ, Ctr Econ Res, Jinan, Shandong, Peoples R China
来源：
JOURNAL OF TIME SERIES ANALYSIS | 2018年 / 39卷 / 02期
关键词：
high-dimensional linear model; square-root LASSO; -mixing; phi-mixing; m-dependent; estimation consistency; CENTRAL-LIMIT-THEOREM; REGRESSION; CONVERGENCE; PREDICTORS; SHRINKAGE; VARIABLES; SELECTION; RECOVERY; MODELS;
D O I：
10.1111/jtsa.12278
中图分类号：
O1 [数学];
学科分类号：
0701 ; 070101 ;
摘要：
We study the square-root LASSO method for high-dimensional sparse linear models with weakly dependent errors. The asymptotic and non-asymptotic bounds for the estimation errors are derived. Our results cover a wide range of weakly dependent errors, including -mixing, -mixing, phi-mixing, and m-dependent types. Numerical simulations are conducted to show the consistency property of square-root LASSO. An empirical application to financial data highlights the importance of the results and method.
引用
收藏
页码：212 / 238
页数：27
相关论文
共 50 条
12345