Dynamic Window-level Granger Causality of Multi-channel Time Series

TL;DR

This paper introduces a dynamic window-level Granger causality method (DWGC) for multi-channel time series, which captures changing causal relationships over time using a novel causality indexing technique.

Contribution

The paper proposes a new DWGC approach with causality indexing that enhances detection of dynamic causalities in multi-channel data, surpassing traditional methods.

Findings

DWGC with causality indexing improves causality detection accuracy.

The method effectively captures dynamic causal relationships.

Experimental results validate the approach on synthetic and real datasets.

Abstract

Granger causality method analyzes the time series causalities without building a complex causality graph. However, the traditional Granger causality method assumes that the causalities lie between time series channels and remain constant, which cannot model the real-world time series data with dynamic causalities along the time series channels. In this paper, we present the dynamic window-level Granger causality method (DWGC) for multi-channel time series data. We build the causality model on the window-level by doing the F-test with the forecasting errors on the sliding windows. We propose the causality indexing trick in our DWGC method to reweight the original time series data. Essentially, the causality indexing is to decrease the auto-correlation and increase the cross-correlation causal effects, which improves the DWGC method. Theoretical analysis and experimental results on two synthetic and one real-world datasets show that the improved DWGC method with causality indexing better detects the window-level causalities.