Causal Regime Detection in Energy Markets With Augmented Time Series Structural Causal Models
Dennis Thumm
TL;DR
The paper tackles the lack of causal interpretability in energy-market modeling by addressing continuous regime changes with ATSCM, a framework that extends augmented structural causal models to multivariate time series. It introduces a three-level generative hierarchy with a learned time-varying graph $\mathcal{G}^t$ and interpretable factors $W^t$, $\\mathbf{I}^t$, and $\\mathbf{V}^t$, enabling counterfactual queries such as $P^*(\\mathbf{V}_{\\tau:T} = \\mathbf{v}' \,|\, \\mathbf{V}^{1:T} = \\mathbf{v}^{1:T}, do(W_{\\tau:T} = w'))$. A neural causal discovery component learns time-varying DAGs without ground-truth graphs, while the training objective combines reconstruction, causal consistency, counterfactual realism, and discovery constraints. Empirical validation on real electricity-market data demonstrates interpretable counterfactual analyses (e.g., under alternative renewable scenarios) and competitive forecasting, with practical implications for scenario planning, policy evaluation, and risk management in energy systems.
Abstract
Energy markets exhibit complex causal relationships between weather patterns, generation technologies, and price formation, with regime changes occurring continuously rather than at discrete break points. Current approaches model electricity prices without explicit causal interpretation or counterfactual reasoning capabilities. We introduce Augmented Time Series Causal Models (ATSCM) for energy markets, extending counterfactual reasoning frameworks to multivariate temporal data with learned causal structure. Our approach models energy systems through interpretable factors (weather, generation mix, demand patterns), rich grid dynamics, and observable market variables. We integrate neural causal discovery to learn time-varying causal graphs without requiring ground truth DAGs. Applied to real-world electricity price data, ATSCM enables novel counterfactual queries such as "What would prices be under different renewable generation scenarios?".