Econometric vs. Causal Structure-Learning for Time-Series Policy Decisions: Evidence from the UK COVID-19 Policies

A research study compares econometric methods versus causal machine learning algorithms for analyzing time-series data to inform policy decisions, using UK COVID-19 policies as a case study. The research evaluates four econometric methods against eleven causal ML algorithms, finding that econometric methods provide clearer temporal structure rules while causal ML algorithms explore broader graph structures to capture more causal relationships.

• →Econometric methods provide clear rules for temporal structures in time-series causal analysis.
• →Causal ML algorithms explore larger spaces of graph structures, leading to denser graphs that capture more identifiable causal relationships.
• →The study bridges econometrics and causal machine learning by providing code to translate econometric results to Bayesian Network libraries.
• →Real-world policy decision-making can benefit from combining insights from both econometric and causal ML approaches.
• →Four econometric methods were evaluated against eleven causal ML algorithms using UK COVID-19 policy data as the test case.