Pathway Lasso: Pathway Estimation and Selection with High-Dimensional Mediators.
Pathway Lasso: Pathway Estimation and Selection with High-Dimensional Mediators.
复制标题
DOI：
10.4310/21-sii673
复制
发表时间：
2022
影响因子：
0.8
通讯作者：
Luo, Xi
中科院分区：
文献类型：
--
作者：
Zhao, Yi;Luo, Xi
在许多科学研究中，通过大量的中介，如遗传和脑中介来描绘这些途径变得越来越重要。结构方程模型是一种常用的估计路径效应的技术，通常表示为系数的乘积。然而，将这样的模型与高维介体相匹配变得不稳定并且在计算上具有挑战性。本文利用正则化结构方程方法提出了一个稀疏中介模型，其中稀疏性是指少数中介在治疗和结果之间具有非零的中介效应。为了解决模型选择的挑战，我们创新地引入了一种名为路径套索的新处罚。该罚函数是对中介效应的非凸积函数的凸松弛，它使得计算上易于处理的优化准则能够同时估计和选择路径效应。我们开发了一个快速的ADMM类型的算法来计算模型参数，并且我们证明了迭代更新可以用闭合的形式来表示。我们还证明了中介效应的路径套索估计的渐近相合性。在模拟数据和fMRI数据集上，该方法比同类方法具有更高的路径选择精度和更低的估计偏差。
In many scientific studies, it becomes increasingly important to delineate the pathways through a large number of mediators, such as genetic and brain mediators. Structural equation modeling (SEM) is a popular technique to estimate the pathway effects, commonly expressed as the product of coefficients. However, it becomes unstable and computationally challenging to fit such models with high-dimensional mediators. This paper proposes a sparse mediation model using a regularized SEM approach, where sparsity means that a small number of mediators have a nonzero mediation effect between a treatment and an outcome. To address the model selection challenge, we innovate by introducing a new penalty called Pathway Lasso. This penalty function is a convex relaxation of the non-convex product function for the mediation effects, and it enables a computationally tractable optimization criterion to estimate and select pathway effects simultaneously. We develop a fast ADMM-type algorithm to compute the model parameters, and we show that the iterative updates can be expressed in closed form. We also prove the asymptotic consistency of our Pathway Lasso estimator for the mediation effect. On both simulated data and an fMRI data set, the proposed approach yields higher pathway selection accuracy and lower estimation bias than competing methods.
登录
查看更多内容
影响因子：
7.6
作者：
BARON, RM;KENNY, DA
通讯作者：
KENNY, DA
影响因子：
1.9
作者：
Daniel RM;De Stavola BL;Cousens SN;Vansteelandt S
通讯作者：
Vansteelandt S
影响因子：
1.9
作者：
Guo R;Zhu H;Chow SM;Ibrahim JG
通讯作者：
Ibrahim JG
影响因子：
5
作者：
Caffo, Brian;Chen, Sining;Schwartz, Brian S.
通讯作者：
Schwartz, Brian S.
影响因子：
1.5
作者：
Bickel, PJ;Levina, E
通讯作者：
Levina, E