Causal Inference from Slowly Varying Nonstationary Processes
This work addresses causal inference problems for researchers dealing with slowly varying nonstationary processes, representing an incremental advancement by adapting existing frameworks to nonstationary settings.
The paper tackled the challenge of causal inference from nonstationary time series by proposing a new class of restricted structural causal models with time-varying filters and stationary noise, demonstrating effectiveness on synthetic and real datasets with high-order and non-smooth filters.
Causal inference from observational data following the restricted structural causal models (SCM) framework hinges largely on the asymmetry between cause and effect from the data generating mechanisms, such as non-Gaussianity or non-linearity. This methodology can be adapted to stationary time series, yet inferring causal relationships from nonstationary time series remains a challenging task. In this work, we propose a new class of restricted SCM, via a time-varying filter and stationary noise, and exploit the asymmetry from nonstationarity for causal identification in both bivariate and network settings. We propose efficient procedures by leveraging powerful estimates of the bivariate evolutionary spectra for slowly varying processes. Various synthetic and real datasets that involve high-order and non-smooth filters are evaluated to demonstrate the effectiveness of our proposed methodology.