Decomposing predictions into between-unit, within-unit-across-time, and counterfactual components shows within-unit accuracy is a structurally better proxy than overall accuracy for recovering true causal treatment effects from non-experimental panel data.
Oscar Barriga-Cabanillas, Joshua E
2 Pith papers cite this work, alongside 9 external citations. Polarity classification is still indexing.
2
Pith papers citing it
9
external citations · OpenAlex
fields
stat.ME 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Proposes an optimal blending framework for proxy and north star metrics in online A/B testing that adjusts decision weights based on statistical power and proxy quality.
citing papers explorer
-
Prediction decomposition for causal analysis
Decomposing predictions into between-unit, within-unit-across-time, and counterfactual components shows within-unit accuracy is a structurally better proxy than overall accuracy for recovering true causal treatment effects from non-experimental panel data.
-
Blending Proxy Metrics with a North Star
Proposes an optimal blending framework for proxy and north star metrics in online A/B testing that adjusts decision weights based on statistical power and proxy quality.