Hyttinen Antti, Plis Sergey, Järvisalo Matti, Eberhardt Frederick, Danks David
HIIT, Department of Computer Science, University of Helsinki.
Mind Research Network and University of New Mexico.
JMLR Workshop Conf Proc. 2016 Aug;52:216-227.
This paper focuses on causal structure estimation from time series data in which measurements are obtained at a coarser timescale than the causal timescale of the underlying system. Previous work has shown that such subsampling can lead to significant errors about the system's causal structure if not properly taken into account. In this paper, we first consider the search for the system timescale causal structures that correspond to a given measurement timescale structure. We provide a constraint satisfaction procedure whose computational performance is several orders of magnitude better than previous approaches. We then consider finite-sample data as input, and propose the first constraint optimization approach for recovering the system timescale causal structure. This algorithm optimally recovers from possible conflicts due to statistical errors. More generally, these advances allow for a robust and non-parametric estimation of system timescale causal structures from subsampled time series data.
本文聚焦于从时间序列数据中进行因果结构估计，其中测量是在比底层系统的因果时间尺度更粗的时间尺度上获得的。先前的工作表明，如果没有适当考虑，这种子采样可能会导致关于系统因果结构的重大误差。在本文中，我们首先考虑寻找与给定测量时间尺度结构相对应的系统时间尺度因果结构。我们提供了一种约束满足程序，其计算性能比以前的方法好几个数量级。然后，我们将有限样本数据作为输入，并提出了第一种用于恢复系统时间尺度因果结构的约束优化方法。该算法能从由于统计误差导致的可能冲突中进行最优恢复。更一般地说，这些进展使得能够从子采样时间序列数据中对系统时间尺度因果结构进行稳健且非参数的估计。
