causal-inference-mixtape

name	causal-inference-mixtape
description	This skill should be used when the user asks to "implement a DiD regression", "run diff-in-discontinuities", "set up an event study", "implement IV / 2SLS", "run a regression discontinuity design", "build synthetic control", "saturated interacted FE", "weather-shock IV", "Callaway-Sant'Anna", "Sun-Abraham", or needs causal-inference code templates in Python/R/Stata with finance applications.
version	2.1.0

Causal Inference: The Mixtape — Code Skill (v2.1)

Provides practitioner-oriented causal-inference templates covering 10 identification strategies in Python, R, and Stata, extended with 25 top-tier Journal of Finance (2021-2024) application case studies. Based on Scott Cunningham's Causal Inference: The Mixtape plus mined JF 2021-2024 top-cited papers for finance-specific patterns.

---

Methods Covered

Method	Python	R	Stata	Reference
OLS / Regression	statsmodels	estimatr	reg/reghdfe	references/method-patterns.md §1
Difference-in-Differences	statsmodels + C()	lfe/fixest	xtreg/reghdfe	references/method-patterns.md §2
Event Study (Dynamic DiD)	manual lead/lag	estimatr	reghdfe	references/method-patterns.md §3
Staggered DiD / TWFE	statsmodels	bacondecomp	bacondecomp	references/method-patterns.md §4
Regression Discontinuity	statsmodels polynomial	rdrobust	rdplot/rdrobust	references/method-patterns.md §5
Instrumental Variables	linearmodels IV2SLS	AER/ivreg	ivregress 2sls	references/method-patterns.md §6
Synthetic Control	rpy2 → R Synth	Synth + SCtools	synth	references/method-patterns.md §7
Matching / PSM / IPW	manual logit + weights	MatchIt + Zelig	teffects/cem	references/method-patterns.md §8
DAGs / Collider Bias	dagitty (conceptual)	dagitty/ggdag	—	references/method-patterns.md §9
Randomization Inference	permutation loop	ri2	ritest	references/method-patterns.md §10

---

Core Workflow

Implement a Causal Method

1. Identify the method from the table above
2. Load the appropriate template from references/method-patterns.md
3. Adapt variable names, fixed effects, and clustering to the user's data
4. Add robustness checks (parallel trends for DiD, McCrary for RDD, first-stage F for IV)

Choose the Right Language

Scenario	Recommendation
ML pipeline integration	Python (statsmodels + linearmodels)
Synthetic Control	R (Synth package) or Stata (synth) — Python lacks mature implementation
Bacon decomposition	R (bacondecomp) or Stata — no Python equivalent
Publication-ready tables	Stata (outreg2/esttab) or R (stargazer/modelsummary)
Coarsened Exact Matching	Stata (cem) or R (MatchIt) — no Python equivalent
Quick prototyping	Python with statsmodels

For cross-language syntax equivalents (OLS, cluster SE, two-way FE, IV, DiD), see references/r-stata-comparison.md §Cross-Language Equivalents.

---

Key Python Patterns

DiD with Cluster-Robust SE

import statsmodels.formula.api as smf

model = smf.ols('y ~ C(treated)*C(post) + controls', data=df)
results = model.fit(cov_type='cluster', cov_kwds={'groups': df['firm_id']})

Event Study (Lead/Lag)

# Create relative time dummies
for k in range(-4, 5):
    col = f'rel_{k}' if k >= 0 else f'rel_m{abs(k)}'
    df[col] = (df['relative_time'] == k).astype(int)

# Drop t=-1 as reference
formula = 'y ~ ' + ' + '.join([c for c in rel_cols if c != 'rel_m1']) + ' + C(id) + C(year)'

IV / 2SLS

from linearmodels.iv import IV2SLS

model = IV2SLS.from_formula('y ~ 1 + exog + [endog ~ instrument]', data=df)
results = model.fit(cov_type='clustered', clusters=df['cluster_var'])

---

Robustness Check Patterns

Method	Required Checks
DiD	Parallel trends (event study plot), placebo treatment dates
RDD	McCrary density test, bandwidth robustness (half/double IK optimal), polynomial robustness
IV	First-stage F > 10, exclusion restriction argument, over-identification test
Synthetic Control	Pre-treatment RMSPE, placebo distribution, leave-one-out
Matching	Covariate balance table, caliper sensitivity

---

Common Pitfalls

1. TWFE with staggered treatment — standard two-way FE is biased when treatment timing varies. Use Bacon decomposition or Sun & Abraham / Callaway & Sant'Anna estimators.
2. Synthetic Control with many treated units — the Synth package handles one treated unit. For multiple, use augmented synthetic control or stacked approach.
3. RDD without McCrary test — always test for manipulation at the cutoff before estimating.
4. IV weak instruments — report first-stage F-statistic. Below 10 indicates weak instrument bias.
5. Python Synth gap — no mature Python Synth package exists. Use rpy2 to call R's Synth from Python.
6. Non-causal framing vs. causal identification — asset-pricing "factor captures the cross-section" is a spanning claim, not a causal one. Do not use "cause / drive / affect" without exogenous variation. See references/identification-writing-patterns.md §3 for verb guide.
7. Saturated FE absorbing treatment — county × year FE absorbs any state-level policy. Always check that FE structure does not soak up the treatment itself.
8. Over-clustering — cluster at the treatment-assignment level (Abadie-Athey-Imbens-Wooldridge 2023), not the highest FE level.
9. Cross-country panel ≠ causal — rich FE reduce confounds but don't produce exogenous variation. Bolton 2023 is honest about this; follow that pattern.
10. Multiple-testing / factor zoo — apply Benjamini-Yekutieli FDR or Bayesian theme-shrinkage (Jensen-Kelly-Pedersen 2023) before trusting any single α.

---

Additional Resources

Reference Files

• references/method-patterns.md — Detailed code templates for all 10 methods with full examples
• references/r-stata-comparison.md — Cross-language package comparison and method coverage gaps
• references/finance-applications.md — 25-paper case-sketch catalog indexed by method (DiD / IV / RDD / event study / structural / Fama-MacBeth) with Y / D / unit / FE / clustering / data-source filled in from top-tier JF papers
• references/identification-writing-patterns.md — Exemplar identification paragraphs, finance-specific verb guide (causal vs. spanning vs. descriptive), and design-specific robustness checklists
• references/jf-case-studies.md — Five end-to-end worked examples (Bennedsen DiD + DDD + diff-in-disc, Kempf saturated FE, Brown weather-shock IV, Barber sharp RDD + outage, Meeuwis election × partisan DiD) with full Python / R / Stata code skeletons

Prompt Files

• prompts/01-implement-method.md — Copy-paste prompt for implementing any causal method
• prompts/02-robustness-checks.md — Copy-paste prompt for generating robustness check code
• prompts/03-finance-application.md — Copy-paste prompt for implementing a causal method on a finance dataset, with robustness checklist and identification-paragraph drafting