Robust Causal Discovery in Real-World Time Series with Power-Laws
Tusoni, Masi, Coletta et al.
Exploring causal relationships in stochastic time series is a challenging yet crucial task with a vast range of applications, including finance, economics, neuroscience, and climate science. Many algorithms for Causal Discovery (CD) have been proposed, but they often exhibit a high sensitivity to noise, resulting in misleading causal inferences when applied to real data. In this paper, we observe that the frequency spectra of typical real-world time series follow a power-law distribution, notably due to an inherent self-organizing behavior. Leveraging this insight, we build a robust CD method based on the extraction of power -law spectral features that amplify genuine causal signals. Our method consistently outperforms state-of-the-art alternatives on both synthetic benchmarks and real-world datasets with known causal structures, demonstrating its robustness and practical relevance.
academic
Robust Causal Discovery in Real-World Time Series with Power-Laws
输入: 时间序列 x = (x₁, ..., xₐ),窗口大小 l,步长 s
输出: 因果图 G
1. 将每个 xᵢ 分割为 ⌊(L-l)/s⌋+1 个滑动窗口 wᵢᵏ
2. for each i ∈ {1, ..., d} do
3. for each k ∈ {0, ..., ⌊(L-l)/s⌋} do
4. 对 wᵢᵏ 应用 DFT 得到 φᵢᵏ
5. 通过方程(2)的拟合得到 (aᵢᵏ, λᵢᵏ)
6. 连接 (aᵢᵏ, λᵢᵏ) 得到时间序列 (aᵢ, λᵢ)
7. for each i,j ∈ {1, ..., d}, i ≠ j do
8. Gᵢ,ⱼ ← Granger因果测试,以(aᵢ,λᵢ)为原因,λⱼ为结果
9. return G