Effect-Level Validation for Causal Discovery

TL;DR

This paper introduces an effect-centric validation framework for causal discovery in telemetry data, emphasizing effect identifiability and stability over mere graph recovery accuracy to ensure reliable decision-making.

Contribution

It proposes a novel admissibility-first approach that evaluates discovered causal graphs based on identifiability, stability, and falsification, improving the reliability of causal inferences in telemetry data.

Findings

Many plausible causal graphs do not support point-identified effects under constraints.

Decision-relevant effect estimates converge across different algorithms when identifiable.

Admissibility-focused validation outperforms structural accuracy in ensuring causal reliability.

Abstract

Causal discovery is increasingly applied to large-scale telemetry data to estimate the effects of user-facing interventions, yet its reliability for decision-making in feedback-driven systems with strong self-selection remains unclear. In this paper, we propose an effect-centric, admissibility-first framework that treats discovered graphs as structural hypotheses and evaluates them by identifiability, stability, and falsification rather than by graph recovery accuracy alone. Empirically, we study the effect of early exposure to competitive gameplay on short-term retention using real-world game telemetry. We find that many statistically plausible discovery outputs do not admit point-identified causal queries once minimal temporal and semantic constraints are enforced, highlighting identifiability as a critical bottleneck for decision support. When identification is possible, several algorithm families converge to similar, decision-consistent effect estimates despite producing substantially different graph structures, including cases where the direct treatment-outcome edge is absent and the effect is preserved through indirect causal pathways. These converging estimates survive placebo, subsampling, and sensitivity refutation. In contrast, other methods exhibit sporadic admissibility and threshold-sensitive or attenuated effects due to endpoint ambiguity. These results suggest that graph-level metrics alone are inadequate proxies for causal reliability for a given target query. Therefore, trustworthy causal conclusions in telemetry-driven systems require prioritizing admissibility and effect-level validation over causal structural recovery alone.