# Everyday Causal Inference

> A practical, code-first guide to causal inference for data scientists, product managers, and analysts. Covers experiments, difference-in-differences, instrumental variables, regression discontinuity, causal time series, and heterogeneous effects — with R and Python code. Written by an economist working in fintech and experimentation.

## Author

- [Robson Tigre](https://www.robsontigre.com), Ph.D. in Economics

## Chapters

### Part I: Foundations

- [Data, Models, and Causality](https://www.everydaycausal.com/data-models-causality.html): ML prediction vs causal inference, OLS regression, estimand-estimator-estimate framework
- [Biases and Causal Frameworks](https://www.everydaycausal.com/causal-frameworks.html): Omitted variable bias, selection bias, potential outcomes, ATE/ATT/ITT/LATE/CATE
- [Planning Your Analysis](https://www.everydaycausal.com/planning-your-analysis.html): Formulating causal questions, avoiding HARKing, study design

### Part II: Estimating impacts

- [Designing Experiments](https://www.everydaycausal.com/designing-experiments.html): SUTVA, randomization, metrics (OEC, driver, guardrail), pre-flight checks
- [Power Analysis](https://www.everydaycausal.com/power-analysis.html): Type I/II errors, minimum detectable effect, sample size calculations
- [Causal Assumptions (DAGs)](https://www.everydaycausal.com/causal-assumptions.html): Frontdoor/backdoor criterion, confounders, mediators, colliders
- [Instrumental Variables](https://www.everydaycausal.com/instrumental-variables.html): Two-stage least squares, exclusion restriction, monotonicity, LATE, weak instruments
- [Regression Discontinuity](https://www.everydaycausal.com/regression-discontinuity.html): Sharp vs fuzzy RDD, running variable, bandwidth, validation tests
- [TWFE Difference-in-Differences](https://www.everydaycausal.com/twfe-did.html): Parallel trends, two-way fixed effects, event studies
- [Staggered DiD](https://www.everydaycausal.com/staggered-did.html): Heterogeneous treatment timing, Callaway and Sant'Anna estimator
- [Time Series Causal Inference](https://www.everydaycausal.com/time-series.html): CausalImpact, CausalArima, synthetic counterfactuals
- [Heterogeneous Effects](https://www.everydaycausal.com/heterogeneous-effects.html): CATE, causal forests, meta-learners

### Part III: Useful, credible estimates, and next steps

- [Falsification and Robustness](https://www.everydaycausal.com/falsification.html): Placebo tests, negative controls, balance and overlap checks, specification grids, sensitivity to unobserved confounding
- [Business Translations](https://www.everydaycausal.com/business-translations.html): Converting causal estimates to ROI, NPV, IRR, payback period, CAC-to-LTV
- [Other Approaches](https://www.everydaycausal.com/other-approaches.html): Structural causal models, causal discovery, causal ML, Bayesian causal inference, time-varying treatments, interference, policy learning, mediation

## Resources

- [Glossary](https://www.everydaycausal.com/glossary.html): Definitions of causal inference terms
- [GitHub Repository](https://github.com/RobsonTigre/everyday-ci): Source code and datasets