causal-inference

Name: causal-inference
Author: Learning-Bayesian-Statistics

153stars15forks

How to install causal-inference?

learning-bayesian-statistics/baygent-skills/causal-inference

$npx -y skills add Learning-Bayesian-Statistics/baygent-skills --skill causal-inference

Installs into the current project.

Summary

Production-grade Bayesian causal inference with PyMC, CausalPy, and DoWhy. Enforces DAG-first thinking, mandatory user checkpoints for assumptions, design-specific refutation, and defensible reporting with causal language guardrails. Trigger on: causal inference, causal effect es

Files · 1

View on GitHub

SKILL.md

1	# Causal Inference
2
3	## Dependencies
4
5	This skill requires the bayesian-workflow skill for all PyMC modeling steps (priors, sampling,
6	diagnostics, calibration, reporting).
7
8	Detect it:
9
10	```bash
11	ls ~/.claude/skills/bayesian-workflow/SKILL.md 2>/dev/null \|\| ls .claude/skills/bayesian-workflow/SKILL.md 2>/dev/null
12	```
13
14	If not found, install it:
15
16	```bash
17	git clone https://github.com/Learning-Bayesian-Statistics/baygent-skills.git /tmp/baygent-skills
18	cp -r /tmp/baygent-skills/bayesian-workflow ~/.claude/skills/
19	```
20
21	For all PyMC modeling steps (priors, sampling, diagnostics, calibration, reporting), follow the
22	bayesian-workflow skill.
23
24	## Workflow overview
25
26	Every causal analysis follows this sequence. Steps 1-4 are the thinking phase (no code). Steps 5-8
27	are the doing phase. Think before you do.
28
29	1. Formulate the causal question — Propose precise estimand (ATE, ATT, LATE, etc.). ⚠️ ASK USER TO CONFIRM.
30	2. Draw the DAG — Propose causal graph with nodes, edges, and explicit non-edges. ⚠️ ASK USER TO CONFIRM. See [references/dags-and-identification.md](references/dags-and-identification.md)
31	3. Identify — Determine identification strategy (backdoor, front-door, IV, RDD, DiD). ⚠️ ASK USER TO CONFIRM untestable assumptions. See [references/dags-and-identification.md](references/dags-and-identification.md)
32	4. Choose design — Match problem to method using table below. ⚠️ ASK USER TO CONFIRM. See [references/quasi-experiments.md](references/quasi-experiments.md) or [references/structural-models.md](references/structural-models.md)
33	5. Estimate — Build and fit the model. Delegate all PyMC mechanics to bayesian-workflow skill.
34	6. Refute — MANDATORY. Run design-specific robustness checks. See [references/refutation.md](references/refutation.md)
35	7. Interpret — Effect size + decision-relevant HDIs + probability of direction.
36	8. Report — Generate `<treatment>-on-<outcome>/report.md` using the canonical template in [references/reporting.md](references/reporting.md). Run `scripts/check_refutation.py` to turn refutation outcomes into pass/marginal/fail ratings, calibrated causal language (causal / suggestive / associational / descriptive), and an ordered next-steps list. Use that output to fill the report's section 7 (causal language calibration) and Suggested Next Steps.
37
38	## Design selection guide
39
40	Before reading the table, walk the intake — it routes you to a design and surfaces the assumptions you'll have to defend later:
41
42	1. Restate the estimand. ATE, ATT, LATE, or a CATE? On whom, over what period? If you can't write the one-sentence estimand, you don't have a causal question yet — you have a dataset.
43	2. Identify the data shape. Single time series, wide panel of units, long unit–time panel, cross-section, or pre/post group data?
44	3. Identify the treatment assignment. Known intervention time, staggered adoption, a threshold/cutoff, a kink, an instrument, or observed treatment with confounders?
45	4. State the identifying story. What makes the comparison causal — parallel trends, no anticipation, no manipulation at the cutoff, a valid instrument, overlap/positivity, donor support? This is exactly what you'll refute later.
46	5. Recommend one primary design + alternatives — and reconcile with the estimand. Name the best-fit design from the table plus any plausible cross-check. Then check that the estimand from step 1 is the one the design actually identifies: DiD → ATT (on the treated); RDD → a local effect at the cutoff; IV → LATE for compliers (under monotonicity); SC → the treated-unit effect. If they differ, change the design or re-scope the estimand and say so. If two designs agree later, that's convergent evidence rather than reliance on a single identification.
47
48	\| Design \| Use when \| Key assumption \| Tool \|
49	\|---\|---\|---\|---\|
50	\| DiD \| Treatment at known time, control group available \| Parallel trends \| CausalPy \|
51	\| Staggered DiD \| Treatment rolls out at different times \| Parallel trends per cohort \| CausalPy \|
52	\| Synthetic Cont

People who liked this also viewed

6 picks

Type

causal-inference

153stars15forks

View on GitHub

How to install causal-inference?

learning-bayesian-statistics/baygent-skills/causal-inference

$npx -y skills add Learning-Bayesian-Statistics/baygent-skills --skill causal-inference

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Causal Inference
2
3	## Dependencies
4
5	This skill requires the bayesian-workflow skill for all PyMC modeling steps (priors, sampling,
6	diagnostics, calibration, reporting).
7
8	Detect it:
9
10	```bash
11	ls ~/.claude/skills/bayesian-workflow/SKILL.md 2>/dev/null \|\| ls .claude/skills/bayesian-workflow/SKILL.md 2>/dev/null
12	```
13
14	If not found, install it:
15
16	```bash
17	git clone https://github.com/Learning-Bayesian-Statistics/baygent-skills.git /tmp/baygent-skills
18	cp -r /tmp/baygent-skills/bayesian-workflow ~/.claude/skills/
19	```
20
21	For all PyMC modeling steps (priors, sampling, diagnostics, calibration, reporting), follow the
22	bayesian-workflow skill.
23
24	## Workflow overview
25
26	Every causal analysis follows this sequence. Steps 1-4 are the thinking phase (no code). Steps 5-8
27	are the doing phase. Think before you do.
28
29	1. Formulate the causal question — Propose precise estimand (ATE, ATT, LATE, etc.). ⚠️ ASK USER TO CONFIRM.
30	2. Draw the DAG — Propose causal graph with nodes, edges, and explicit non-edges. ⚠️ ASK USER TO CONFIRM. See [references/dags-and-identification.md](references/dags-and-identification.md)
31	3. Identify — Determine identification strategy (backdoor, front-door, IV, RDD, DiD). ⚠️ ASK USER TO CONFIRM untestable assumptions. See [references/dags-and-identification.md](references/dags-and-identification.md)
32	4. Choose design — Match problem to method using table below. ⚠️ ASK USER TO CONFIRM. See [references/quasi-experiments.md](references/quasi-experiments.md) or [references/structural-models.md](references/structural-models.md)
33	5. Estimate — Build and fit the model. Delegate all PyMC mechanics to bayesian-workflow skill.
34	6. Refute — MANDATORY. Run design-specific robustness checks. See [references/refutation.md](references/refutation.md)
35	7. Interpret — Effect size + decision-relevant HDIs + probability of direction.
36	8. Report — Generate `<treatment>-on-<outcome>/report.md` using the canonical template in [references/reporting.md](references/reporting.md). Run `scripts/check_refutation.py` to turn refutation outcomes into pass/marginal/fail ratings, calibrated causal language (causal / suggestive / associational / descriptive), and an ordered next-steps list. Use that output to fill the report's section 7 (causal language calibration) and Suggested Next Steps.
37
38	## Design selection guide
39
40	Before reading the table, walk the intake — it routes you to a design and surfaces the assumptions you'll have to defend later:
41
42	1. Restate the estimand. ATE, ATT, LATE, or a CATE? On whom, over what period? If you can't write the one-sentence estimand, you don't have a causal question yet — you have a dataset.
43	2. Identify the data shape. Single time series, wide panel of units, long unit–time panel, cross-section, or pre/post group data?
44	3. Identify the treatment assignment. Known intervention time, staggered adoption, a threshold/cutoff, a kink, an instrument, or observed treatment with confounders?
45	4. State the identifying story. What makes the comparison causal — parallel trends, no anticipation, no manipulation at the cutoff, a valid instrument, overlap/positivity, donor support? This is exactly what you'll refute later.
46	5. Recommend one primary design + alternatives — and reconcile with the estimand. Name the best-fit design from the table plus any plausible cross-check. Then check that the estimand from step 1 is the one the design actually identifies: DiD → ATT (on the treated); RDD → a local effect at the cutoff; IV → LATE for compliers (under monotonicity); SC → the treated-unit effect. If they differ, change the design or re-scope the estimand and say so. If two designs agree later, that's convergent evidence rather than reliance on a single identification.
47
48	\| Design \| Use when \| Key assumption \| Tool \|
49	\|---\|---\|---\|---\|
50	\| DiD \| Treatment at known time, control group available \| Parallel trends \| CausalPy \|
51	\| Staggered DiD \| Treatment rolls out at different times \| Parallel trends per cohort \| CausalPy \|
52	\| Synthetic Cont

People who liked this also viewed

6 picks

Type