Causal Discovery in Semi-Stationary Time Series

TL;DR

This paper introduces PCMCI$_{ ext{Ω}}$, a non-parametric, constraint-based algorithm designed to uncover causal relations in semi-stationary time series that exhibit periodic changes in causal mechanisms, addressing a key challenge in non-stationary data analysis.

Contribution

The paper presents a novel algorithm for causal discovery in semi-stationary time series, capable of handling recurring causal mechanism changes without assuming stationarity.

Findings

The algorithm accurately identifies causal relations in simulated data.

It effectively captures periodic causal mechanism changes.

Applied to climate data, it reveals meaningful causal structures.

Abstract

Discovering causal relations from observational time series without making the stationary assumption is a significant challenge. In practice, this challenge is common in many areas, such as retail sales, transportation systems, and medical science. Here, we consider this problem for a class of non-stationary time series. The structural causal model (SCM) of this type of time series, called the semi-stationary time series, exhibits that a finite number of different causal mechanisms occur sequentially and periodically across time. This model holds considerable practical utility because it can represent periodicity, including common occurrences such as seasonality and diurnal variation. We propose a constraint-based, non-parametric algorithm for discovering causal relations in this setting. The resulting algorithm, PCMCI$_{\Omega}$, can capture the alternating and recurring changes in the causal mechanisms and then identify the underlying causal graph with conditional independence (CI) tests. We show that this algorithm is sound in identifying causal relations on discrete time series. We validate the algorithm with extensive experiments on continuous and discrete simulated data. We also apply our algorithm to a real-world climate dataset.