CAM: CAUSAL ADDITIVE MODELS, HIGH-DIMENSIONAL ORDER SEARCH AND PENALIZED REGRESSION
CAM: CAUSAL ADDITIVE MODELS, HIGH-DIMENSIONAL ORDER SEARCH AND PENALIZED REGRESSION
复制标题
DOI：
10.1214/14-aos1260
复制
发表时间：
2014-12-01
影响因子：
4.5
通讯作者：
Ernest, Jan
中科院分区：
文献类型：
--
作者：
Buehlmann, Peter;Peters, Jonas;Ernest, Jan
我们开发潜在的高维加性结构方程模型的估计。我们的方法的一个关键组成部分是解耦顺序搜索的变量之间的特征或边缘选择的有向无环图编码的因果结构。我们表明，前者可以通过非正则化（限制）最大似然估计来完成，而后者可以使用稀疏回归技术来有效地解决。因此，我们大大简化了一类重要的因果模型的结构搜索和估计问题。我们建立一致性的（限制）最大似然估计的低维和高维的情况下，我们也允许错误的误差分布：此外，我们开发了一个有效的计算算法，可以处理许多变量，和新方法的准确性和性能的模拟和真实的数据。
We develop estimation for potentially high-dimensional additive structural equation models. A key component of our approach is to decouple order search among the variables from feature or edge selection in a directed acyclic graph encoding the causal structure. We show that the former can be done with nonregularized (restricted) maximum likelihood estimation while the latter can be efficiently addressed using sparse regression techniques. Thus, we substantially simplify the problem of structure search and estimation for an important class of causal models. We establish consistency of the (restricted) maximum likelihood estimator for low- and high-dimensional scenarios, and we also allow for misspecification of the error distribution: Furthermore, we develop an efficient computational algorithm which can deal with many variables, and the new method's accuracy and performance is illustrated on simulated and real data.