# Empirical Design

> Empirical Design

- Skill: `chanw-research/empirical-design` (Agent Skill)
- Install (CLI): `npx skillmds@latest add chanw-research/empirical-design`
- Raw SKILL.md: https://api.skillmd.com/api/skills/chanw-research/empirical-design/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: chanw-research (https://skillmd.com/u/chanw-research)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/chanw-research/empirical-design

---


# /empirical-design

Read the wiki context, empirical pages, project READMEs, and `raw/notes/research-intent.md` when present.

Generate an archived design document at `wiki/outputs/empirical-design-{slug}-{YYYY-MM-DD}.md` with:

1. Research question
2. Literature positioning
3. Theory mechanism
4. Hypotheses
5. Variable design
6. Data sources and sample construction
7. Baseline model
8. Identification strategy and endogeneity risks
9. Mechanism tests
10. Heterogeneity tests
11. Robustness checks
12. Expected table structure
13. Data gaps
14. Next actions

Constraints:

- Distinguish what the literature already did from what this project should do.
- Be concrete about endogeneity risks and feasible remedies.
- If local data are insufficient, list the missing fields rather than pretending the design is executable.