Causal Inference Assumptions: A Practical Checklist Before You Run the Model
Nasereliver · x · 2026-09-30
Data scientist raheem nasirudeen published a tutorial-style blog arguing that causal inference requires thinking before modeling — "data itself lacks intelligence."
Key points:
- Starts from RCTs as the gold standard and frames the goal as identification: making observational data resemble RCT conditions
- Distinguishes cross-sectional, panel, and time-varying confounding data types
- Systematically walks through assumptions across the Rubin and Pearl frameworks: unconfoundedness, the backdoor criterion, good vs. bad controls, positivity, modularity, SUTVA (including spillover), consistency, parallel trends, and no anticipation
- Aimed at practitioners estimating treatment effects from observational data, urging them to verify assumptions before running any model
More from Research
- Fully open-source VLA MolmoAct2 lands CoRL 2026 Spotlight, matching π0.5 performance — jmin__cho · 2026-09-30
- Image Lab Tops Robot Leaderboard: FLUX 3 Action Runs on a Desk-Size Arm — lmoroney · 2026-09-30
- MIT builds AI-powered Raman 'barcode' to noninvasively detect senescent 'zombie cells' — MacrinePhD · 2026-09-30
- Anil Seth's biological naturalism essay against AI consciousness draws mass commentary collection — anilkseth · 2026-09-30
- Sparse-View Gaussian Splatting Demo Shows 3D Reconstruction From Few Images — enndeeee · 2026-09-30
- Open-source self-driving guide adds VFH: binning LiDAR into 5° obstacle histograms — 4310sy · 2026-09-30