Counterfactual Outcomes Lab Rubin Causal Model

Estimating unobserved baselines Y(0) via Synthetic Control Methods
Causal Effect (τ)
14.50
Mean post-treatment delta
Cumulative Delta
116.0
Total averted / excess events
Pre-Treatment RMSE
0.42
Fit accuracy before T₀
Parallel Trends Test
p = 0.88
Null: Equal pre-trends

Unit Trajectories & Synthetic Baseline

Observed Y₁(t) Synthetic Y₀(t) Controls

Fundamental Problem of Causal Inference

We cannot observe both Y₁(t) and Y₀(t) for the same unit simultaneously after intervention period T₀. The synthetic control method constructs the unobserved counterfactual Y₀(t) using a weighted combination of unexposed control cohorts.

Proof & Telemetry State

Initializing telemetry...

Intervention Setup

Intervention Time (T₀) Period 12
Treatment Intensity +2.5
Uncertainty Level (σ) 0.5

Control Weights (wk) Sum: 1.00

Diagnostics & Assumptions

Rubin Model: τt = Y1,t - Y0,t
Pre-period Fit: RMSE across t < T₀
SUTVA: Assumed no spillover between units.
Enjoy this tool? Build your own with Super