Inference in High-Dimensional Regression Models without the Exact or Lp sparsity

• 成果类型: Article
• 署名作者: Cha, Jooyoung; Chiang, Harold D.; Sasaki, Yuya
• 署名单位: Vanderbilt University; University of Wisconsin System; University of Wisconsin Madison
• 刊物名称: REVIEW OF ECONOMICS AND STATISTICS
• ISSN/ISSBN: 0034-6535
• DOI: 10.1162/rest_a_01349
• 发表日期: 2025
• 关键词: confidence-intervals; selection; parameters; regions; errors; inputs

摘要

We propose a new inference method in high-dimensional regression models and high-dimensional IV regression models. The method is shown to be valid without requiring the exact sparsity or L-p sparsity conditions. Simulation studies demonstrate superior performance of this proposed method over those based on LASSO or random forest, especially under less sparse models. We illustrate an application to production analysis with a panel of Chilean firms.