arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1202.3779cs.LGstat.ML
收藏

检验线性方程是否是因果关系:一种自由概率论方法

Testing whether linear equations are causal: A free probability theory approach

Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);Max Planck Institute for Biogeochemistry(马克斯·普朗克生物地球化学研究所)

Jakob Zscheischler, Dominik Janzing, Kun Zhang

摘要:

我们提出了一种方法,推断两个高维变量X和Y之间的线性关系是否是由于X到Y或从Y到X的因果影响。早期提出的所谓的跟踪方法被扩展到政权的尺寸的观察变量超过样本大小。基于以前的工作,我们假设的条件,表征X和Y之间的因果关系。此外,我们描述了一个统计测试,并认为,这两个因果方向通常被拒绝,如果有一个共同的原因。一个完整的理论分析的确定性的情况下,但我们的方法似乎是有效的噪声的情况下,也是,我们另外提出了一种基于稀疏约束的方法。所讨论的方法产生有前途的结果模拟和真实世界的数据。

英文摘要:

We propose a method that infers whether linear relations between two high-dimensional variables X and Y are due to a causal influence from X to Y or from Y to X. The earlier proposed so-called Trace Method is extended to the regime where the dimension of the observed variables exceeds the sample size. Based on previous work, we postulate conditions that characterize a causal relation between X and Y. Moreover, we describe a statistical test and argue that both causal directions are typically rejected if there is a common cause. A full theoretical analysis is presented for the deterministic case but our approach seems to be valid for the noisy case, too, for which we additionally present an approach based on a sparsity constraint. The discussed method yields promising results for both simulated and real world data.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑