Causal Inference — Name the Assumption, Estimate, Then Try to Break It
Skill type: ANALYSIS MODULE. Estimates a causal effect from data that was not fully
randomized. The discipline is not the estimator — it is the identifying assumption the
estimator relies on, stated before the fit and stress-tested after. If the design is not yet
fixed, that belongs upstream in alterlab-ssci-design-gate.
Core Mission
EVERY CAUSAL ESTIMATE INHERITS AN ASSUMPTION. STATE IT, ESTIMATE UNDER IT, THEN REFUTE IT.
When to Use This Skill
- "Estimate the effect with difference-in-differences / an event study."
- "I have an instrument for the treatment — run instrumental variables / 2SLS."
- "There's a cutoff score — run a regression discontinuity."
- "Panel data with unit and time fixed effects — estimate the treatment effect."
- "Give me the heterogeneous treatment effect / CATE across subgroups."
Does NOT Trigger
| The request is really about… |
Route to |
Why not this skill |
| Choosing the design & its identifying assumption |
alterlab-ssci-design-gate |
Design routing precedes estimation. |
| Plain OLS / descriptive or inferential stats (no causal identification) |
alterlab-statistical-analysis / alterlab-statsmodels |
No treatment-effect identification problem. |
| Latent-variable / SEM / factor structure |
alterlab-sem-psychometrics |
Measurement model, not treatment effect. |
| Auditing whether the final claim is licensed |
alterlab-ssci-inference-gate |
Claim audit, downstream. |
Estimator map (verified libraries, pinned)
| Design |
Identifying assumption |
Estimator (verified API) |
| DiD / event study |
parallel trends |
pyfixest (v0.60): `pf.feols("y ~ i(time, treat, ref) |
| Panel fixed effects |
no time-varying confounders |
linearmodels (v7): PanelOLS.from_formula("y ~ 1 + x + EntityEffects + TimeEffects", panel).fit(cov_type='clustered', cluster_entity=True). |
| Instrumental variables |
exclusion restriction + relevance |
linearmodels IV2SLS.from_formula("y ~ 1 + exog + [treat ~ z1 + z2]", df).fit(); check first-stage F (weak instrument). |
| Regression discontinuity |
continuity at the cutoff (no sorting) |
rdrobust (v2): rdrobust(y, x, c=cutoff), rdbwselect, rdplot; McCrary/density check for manipulation. |
| Selection-on-observables |
conditional ignorability |
DoWhy (v0.14): CausalModel(df, treatment, outcome, graph).identify_effect() → estimate_effect(method_name="backdoor.propensity_score_matching") → refute_estimate(..., method_name="random_common_cause"). |
| Heterogeneous effects (CATE) |
(as above) + overlap |
EconML (v0.16): LinearDML() / CausalForestDML() .fit(Y, T, X=X, W=W) then .effect(X). |
A stdlib router that maps the design to the estimator + its assumption + the verified call:
scripts/estimator_router.py. Fuller worked patterns and diagnostics:
references/estimators.md.
The mandatory two steps around every estimate
- Before: write the identifying assumption in one sentence and how you will defend it
(pre-trend plot for DiD, first-stage F and an exclusion argument for IV, density/McCrary test
for RDD, overlap/common-support for PSM). No assumption, no causal estimate.
- After: run a refutation/robustness check —
DoWhy.refute_estimate (random common cause,
placebo treatment, data subset), a pre-trend/event-study plot, a bandwidth-sensitivity for RDD,
or an E-value / sensitivity analysis for unmeasured confounding. Report it next to the estimate.
Output Template
DESIGN + ASSUMPTION: <e.g. DiD; parallel trends, defended by the pre-2019 event-study plot>
ESTIMATOR: <library.call(...), pinned version>
ESTIMATE: <point estimate, 95% CI, clustered SE — never a bare p-value>
DIAGNOSTIC: <first-stage F / density test / pre-trends / overlap>
REFUTATION: <placebo / random-common-cause / bandwidth sensitivity result>
CLAIM SCOPE: causal IFF the assumption + diagnostics hold; else associational
Quality Standards
- Report effect sizes with confidence intervals and the SE structure (clustered where relevant).
- Never present a causal estimate without its diagnostic and at least one refutation.
- Hand the final estimate + assumption to
alterlab-ssci-inference-gate for the claim audit.
- No fabricated flags: every call above is verified against the library's current docs; if a
flag is unverified in your installed version, check
--help/docs, do not guess.
References
references/estimators.md — per-design worked calls, diagnostics, and refutation menu.scripts/estimator_router.py — stdlib design→estimator+assumption router.
Part of the AlterLab Academic Skills suite.
1---2name: alterlab-causal-inference3description: Estimates causal effects from observational and quasi-experimental data — difference-in-differences, instrumental variables, regression discontinuity, panel fixed effects, propensity-score / doubly-robust methods, and heterogeneous treatment effects (CATE) — using the verified Python stack: statsmodels and linearmodels (PanelOLS, IV2SLS), pyfixest (feols, event studies, Sun-Abraham, did2s), DoWhy (identify -> estimate -> refute), EconML (LinearDML, CausalForestDML, DRLearner), and rdrobust for RD. It names the identifying assumption before estimating and runs a refutation/robustness check after. Use when the request mentions difference-in-differences, instrumental variables, regression discontinuity, fixed effects / panel causal estimation, propensity scores, or treatment-effect estimation from non-randomized data. For choosing the design first prefer alterlab-ssci-design-gate; for plain regression or descriptive stats prefer alterlab-statistical-analysis. Part of the AlterLab Academic Skills suite.4license: MIT5---67# Causal Inference — Name the Assumption, Estimate, Then Try to Break It89**Skill type: ANALYSIS MODULE.** Estimates a causal effect from data that was not fully10randomized. The discipline is not the estimator — it is the **identifying assumption** the11estimator relies on, stated before the fit and stress-tested after. If the design is not yet12fixed, that belongs upstream in `alterlab-ssci-design-gate`.1314## Core Mission1516```17EVERY CAUSAL ESTIMATE INHERITS AN ASSUMPTION. STATE IT, ESTIMATE UNDER IT, THEN REFUTE IT.18```1920## When to Use This Skill2122- "Estimate the effect with difference-in-differences / an event study."23- "I have an instrument for the treatment — run instrumental variables / 2SLS."24- "There's a cutoff score — run a regression discontinuity."25- "Panel data with unit and time fixed effects — estimate the treatment effect."26- "Give me the heterogeneous treatment effect / CATE across subgroups."2728### Does NOT Trigger2930| The request is really about… | Route to | Why not this skill |31|---|---|---|32| Choosing the design & its identifying assumption | `alterlab-ssci-design-gate` | Design routing precedes estimation. |33| Plain OLS / descriptive or inferential stats (no causal identification) | `alterlab-statistical-analysis` / `alterlab-statsmodels` | No treatment-effect identification problem. |34| Latent-variable / SEM / factor structure | `alterlab-sem-psychometrics` | Measurement model, not treatment effect. |35| Auditing whether the final claim is licensed | `alterlab-ssci-inference-gate` | Claim audit, downstream. |3637## Estimator map (verified libraries, pinned)3839| Design | Identifying assumption | Estimator (verified API) |40|--------|------------------------|--------------------------|41| **DiD / event study** | parallel trends | `pyfixest` (v0.60): `pf.feols("y ~ i(time, treat, ref) | unit + time", df)`, `pf.event_study(...)`, `pf.did2s(...)`; or `statsmodels` `smf.ols("y ~ treat*post").fit(cov_type='cluster', cov_kwds={'groups': df.unit})`. Staggered adoption → Sun-Abraham `sunab()` or `did2s`. |42| **Panel fixed effects** | no time-varying confounders | `linearmodels` (v7): `PanelOLS.from_formula("y ~ 1 + x + EntityEffects + TimeEffects", panel).fit(cov_type='clustered', cluster_entity=True)`. |43| **Instrumental variables** | exclusion restriction + relevance | `linearmodels` `IV2SLS.from_formula("y ~ 1 + exog + [treat ~ z1 + z2]", df).fit()`; check first-stage F (weak instrument). |44| **Regression discontinuity** | continuity at the cutoff (no sorting) | `rdrobust` (v2): `rdrobust(y, x, c=cutoff)`, `rdbwselect`, `rdplot`; McCrary/density check for manipulation. |45| **Selection-on-observables** | conditional ignorability | `DoWhy` (v0.14): `CausalModel(df, treatment, outcome, graph).identify_effect()` → `estimate_effect(method_name="backdoor.propensity_score_matching")` → `refute_estimate(..., method_name="random_common_cause")`. |46| **Heterogeneous effects (CATE)** | (as above) + overlap | `EconML` (v0.16): `LinearDML()` / `CausalForestDML()` `.fit(Y, T, X=X, W=W)` then `.effect(X)`. |4748A stdlib router that maps the design to the estimator + its assumption + the verified call:49`scripts/estimator_router.py`. Fuller worked patterns and diagnostics:50`references/estimators.md`.5152## The mandatory two steps around every estimate53541. **Before**: write the identifying assumption in one sentence and how you will defend it55 (pre-trend plot for DiD, first-stage F and an exclusion argument for IV, density/McCrary test56 for RDD, overlap/common-support for PSM). No assumption, no causal estimate.572. **After**: run a refutation/robustness check — `DoWhy.refute_estimate` (random common cause,58 placebo treatment, data subset), a pre-trend/event-study plot, a bandwidth-sensitivity for RDD,59 or an E-value / sensitivity analysis for unmeasured confounding. Report it next to the estimate.6061## Output Template6263```64DESIGN + ASSUMPTION: <e.g. DiD; parallel trends, defended by the pre-2019 event-study plot>65ESTIMATOR: <library.call(...), pinned version>66ESTIMATE: <point estimate, 95% CI, clustered SE — never a bare p-value>67DIAGNOSTIC: <first-stage F / density test / pre-trends / overlap>68REFUTATION: <placebo / random-common-cause / bandwidth sensitivity result>69CLAIM SCOPE: causal IFF the assumption + diagnostics hold; else associational70```7172## Quality Standards7374- Report effect sizes with confidence intervals and the SE structure (clustered where relevant).75- Never present a causal estimate without its diagnostic and at least one refutation.76- Hand the final estimate + assumption to `alterlab-ssci-inference-gate` for the claim audit.77- No fabricated flags: every call above is verified against the library's current docs; if a78 flag is unverified in your installed version, check `--help`/docs, do not guess.7980## References8182- `references/estimators.md` — per-design worked calls, diagnostics, and refutation menu.83- `scripts/estimator_router.py` — stdlib design→estimator+assumption router.8485Part of the AlterLab Academic Skills suite.
AlterLab-IEU