Econometrics
Routing Summary
This folder covers applied econometrics and causal inference from Mostly Harmless Econometrics plus Bayesian DiD, synthetic control, DAG tutorials, Bayesian propensity weighting, classical propensity score matching (PSM), simulation-based estimation, staggered difference-in-differences, high-dimensional dependence (copula) modelling, and quasi-Bayesian GMM under plausible (non-exact) moment conditions. Contains 53 notes across 7 sub-topics.
- Need research design fundamentals, selection bias, or DAGs? → Foundations
- Need regression interpretation or OVB? → Regression Foundations
- Need IV, DiD (canonical), RD, synthetic control, GSC, DAGs, or propensity weighting? → Identification Strategies
- Need staggered/multi-period DiD (group-time ATT, doubly-robust estimands, event-study aggregation, multiplier-bootstrap inference)? → Difference-in-Differences
- Need synthetic DiD, event studies, pre-trend testing or Honest DiD? → Synthetic Difference-in-Differences
- Need causal machine learning (double/debiased ML, Neyman orthogonality, cross-fitting, causal forests / GRF, R-learner)? → Causal Machine Learning
- Need quantile regression, discrete choice, or SEs? → Extensions
- Need simulation-based estimation (MSM, indirect inference, EMM, copula SMM)? → Extensions
- Need high-dimensional dependence / copulas, tail dependence, or factor copulas? → Dependence Modeling
- Need quasi-Bayesian GMM with plausible (non-exact) moment conditions, priors over misspecification, or the “no free lunch” weighting trade-off? → Plausible GMM
Book Overview
- Mostly Harmless Econometrics - Overview — Master index for the book’s concepts and structure (moved to Identification Strategies/)
Sub-topics
| Sub-topic | Notes | Domain |
|---|---|---|
| Foundations | 4 | Research questions, experimental ideal, selection bias, DAGs (MHE Part I + Pearl) |
| Regression Foundations | 3 | CEF, CIA, omitted variables bias (MHE Ch 3) |
| Identification Strategies | 27 | IV, LATE, DD, RD, synthetic control, GSC, DAGs, Bayesian IPTW, classical PSM (Rosenbaum-Rubin matching framework + diagnostics) — quasi-experimental methods (MHE Ch 4-6 + Abadie 2021 + Xu 2017 + extras); Synthetic DiD sub-topic (8 notes: SDID estimator and inference, geo-experiment application, event studies, pre-trend testing, Honest DiD) |
| Difference-in-Differences | 6 | Staggered/multi-period DiD: group-time ATT(g,t), parallel-trends/no-anticipation/overlap assumptions, OR/IPW/doubly-robust estimands, event-study/group/calendar aggregation, multiplier-bootstrap uniform inference, TWFE critique (Callaway & Sant’Anna 2020) |
| Extensions | 23 | Quantile regression, discrete choice, standard errors (MHE Ch 7-8), simulation-based estimation: MSM, indirect inference, EMM, SMM for copulas, and foundational time-series SME theory (Liesenfeld & Breitung 1998, Evans 2024, Oh & Patton 2011, Duffie & Singleton 1993) |
| Dependence Modeling | 6 | High-dimensional copulas, factor-copula construction, tail dependence via EVT, multi-factor/block structures, rank-based SMM, S&P 100 & systemic risk (Oh & Patton 2012) |
| Plausible GMM | 5 | Quasi-Bayesian inference when moment conditions are plausible but not exact: plausibility characteristic , proper prior over misspecification, CU-GMM quasi-posterior, local Gaussian prior approximation & “no free lunch”, institutions-and-GDP IV application (Chernozhukov, Hansen, Kong & Wang 2026) |
| Causal Machine Learning | 9 | Regularization bias and the partially linear model, Neyman orthogonality, cross-fitting (DML1/DML2), DML for ATE / ATTE / LATE with AIPW scores, honest trees and causal forests, generalized random forests (local moment equations), forest asymptotics and inference, R-learner — Chernozhukov et al. 2018, Wager & Athey 2018, Athey-Tibshirani-Wager 2019, Nie & Wager 2021 |
Sources
- Mostly Harmless Econometrics — Full textbook PDF (Angrist & Pischke, 2008)
- Discrete Choice and Random Utility Models — PyMC tutorial: Bayesian discrete choice models (McFadden framework)
- Difference in differences — PyMC tutorial: Bayesian DiD with counterfactual prediction
- 15 - Synthetic Control — Causal Inference for the Brave and True — Causal Inference for the Brave and True, Ch. 15: synthetic control with Python (scipy, sklearn)
- Abadie 2021 - Using Synthetic Controls — Abadie (2021) JEL: authoritative guide to synthetic controls, bias theory, requirements, extensions
- Xu 2016 - Generalized Synthetic Control Method — Xu (2017) Political Analysis: GSC method unifying DID and SC via IFE model
- Unlock the Secrets of Causal Inference with a Master Class in Directed Acyclic Graphs — Graham Harrison, Towards Data Science (2023): comprehensive DAG tutorial
- How to use Bayesian propensity scores and inverse probability weights — Andrew Heiss (2021-12-18): Liao-Zigler Bayesian IPW in R/brms
- tdb136 — Liesenfeld & Breitung (1998), “Simulation Based Methods of Moments in Empirical Finance”
- Oh_Patton_SMM_copulas_nov11 — Oh & Patton (2011), “Simulated Method of Moments Estimation for Copula-Based Multivariate Models”
- 19 — Evans (2024), CompMethods Ch. 19: full Python SMM tutorial + Brock-Mirman structural macro exercise
- Oh-Patton-2012-Factor-Copulas — Oh & Patton (2012), “Modelling Dependence in High Dimensions with Factor Copulas” (Duke): factor copulas, EVT tail dependence, rank-based SMM, S&P 100 systemic risk
- 1803.09015-Callaway-SantAnna-DiD-Multiple-Periods — Callaway & Sant’Anna (2020), “Difference-in-Differences with Multiple Time Periods” (J. Econometrics): group-time ATT, doubly-robust estimands, aggregation schemes, multiplier-bootstrap inference, minimum-wage application
- Plausible GMM - A Quasi-Bayesian Approach — Chernozhukov, Hansen, Kong & Wang (2026), arXiv:2507.00555 (econ.EM): quasi-Bayesian GMM under plausible (non-exact) moment conditions, priors over misspecification, Bernstein–von Mises concentration, institutions-and-GDP IV application
- Duffie Singleton 1993 - Simulated Moments Estimation of Markov Models of Asset Prices — Duffie & Singleton (1993), Econometrica 61(4):929–952: foundational Simulated Moments Estimator (SME) theory for time-series Markov asset-pricing models — geometric ergodicity, AUC condition, consistency, asymptotic normality
- PSM-Rosenbaum-Rubin-Stuart-Survey — Survey synthesis: Rosenbaum & Rubin (1983) Biometrika, Rosenbaum & Rubin (1985) AmStat, Stuart (2010) Statistical Science, Imbens (2004) RESTAT — classical propensity score matching framework, algorithms, and balance diagnostics
See Also
- Bayesian Statistics — Bayesian perspective on regression and inference
- Research Methodology — Multiple comparisons and causal inference challenges