Econometric methods impose clear temporal rules on causal structures from time series, whereas causal ML algorithms produce denser graphs that recover more identifiable causal effects in UK COVID-19 policy data.
(1969) Fitting autoregressive models for prediction
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Evidence from real and synthetic time series indicates that frequency-domain AIC bandwidth selection for nonparametric spectral estimation yields results comparable to standard parametric AR-AIC.
citing papers explorer
-
Econometric vs. Causal Structure-Learning for Time-Series Policy Decisions: Evidence from the UK COVID-19 Policies
Econometric methods impose clear temporal rules on causal structures from time series, whereas causal ML algorithms produce denser graphs that recover more identifiable causal effects in UK COVID-19 policy data.
-
Bandwidth selection with a frequency-domain version of the AIC
Evidence from real and synthetic time series indicates that frequency-domain AIC bandwidth selection for nonparametric spectral estimation yields results comparable to standard parametric AR-AIC.