Institute of Biomedicine and Molecular Immunology, National Research Council, Palermo, Italy.
Dipartimento di Scienze Economiche, Aziendali e Statistiche, Università degli Studi di Palermo, Palermo, Italy.
Stat Methods Med Res. 2020 Mar;29(3):765-777. doi: 10.1177/0962280219842890. Epub 2019 Apr 16.
This paper focuses on hypothesis testing in lasso regression, when one is interested in judging statistical significance for the regression coefficients in the regression equation involving a lot of covariates. To get reliable -values, we propose a new lasso-type estimator relying on the idea of induced smoothing which allows to obtain appropriate covariance matrix and Wald statistic relatively easily. Some simulation experiments reveal that our approach exhibits good performance when contrasted with the recent inferential tools in the lasso framework. Two real data analyses are presented to illustrate the proposed framework in practice.
本文专注于套索回归中的假设检验，当人们对涉及大量协变量的回归方程中的回归系数的统计显著性感兴趣时。为了获得可靠的 p 值，我们提出了一种新的基于诱导平滑思想的套索型估计量，该估计量允许相对容易地获得适当的协方差矩阵和 Wald 统计量。一些模拟实验表明，与套索框架中的最近的推断工具相比，我们的方法具有良好的性能。通过两个实际数据分析来说明所提出的框架在实践中的应用。
