On the validity of the pairs bootstrap for lasso estimators

• 成果类型: Article
• 署名作者: Camponovo, L.
• 署名单位: University of St Gallen
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asv039
• 发表日期: 2015
• 页码: 981987
• 关键词: adaptive lasso; models

摘要

We study the validity of the pairs bootstrap for lasso estimators in linear regression models with random covariates and heteroscedastic error terms. We show that the naive pairs bootstrap does not provide a valid method for approximating the distribution of the lasso estimator. To overcome this deficiency, we introduce a modified pairs bootstrap procedure and prove its consistency. Finally, we consider the adaptive lasso and show that the modified pairs bootstrap consistently estimates the distribution of the adaptive lasso estimator.