AI & Economics Research
I created Economics Topic Scout, an AI research skill that helps turn a broad research interest into a feasible economics paper idea.
Research often begins with a broad interest rather than a fully formed question. We explore, follow promising leads, and revise our direction until we find a question that interests us, addresses a gap in the literature, and can be studied with available data and a credible research design.
AI can take on much of the literature search, data checking, and preliminary analysis, lowering the cost of trying an idea, learning from it, and trying again. My skill organizes this iterative search; the 3D map below makes the process visible.
From IO & Health to ETC
My search began with industrial organization and health economics and led to Medicare’s End-Stage Renal Disease Treatment Choices (ETC) Model. Follow the arrows through the questions, branches, and revisions that shaped the paper.
Drag to rotate. Select a node to read its story.
View controls
Use the buttons below, pinch, or hold Ctrl / ⌘ while scrolling to zoom.
- The check
- What followed
- The lesson
About this research record
This account condenses the supplied research map while retaining all 12 main stages and 119 branches. It is reconstructed from research notes and a manuscript snapshot through October 3, 2026. The period before August 4 is based on my recollection.
Arrows show the direction of the research process, not causal relationships or equal units of time. Earlier branches retain their historical judgments. Estimates from different data versions and inference rules should not be combined; a promising exploratory lead is not a confirmed result.
Use the workflow in your own research
Start with a broad interest
Name a market, behavior, or policy that interests you. You do not need a finished research question. Use the skill to explore several possible directions.
Explore questions and gaps
Follow the questions that become interesting as you learn. Read the closest literature, compare alternative ideas, and identify what each could add to what is already known.
Test whether an idea is feasible
Inspect actual data, check the institutional setting and research design, and run focused preliminary tests. AI can help with much of this work; the researcher evaluates the evidence and decides what to pursue.
Revise, branch, and develop
Keep a record of what worked, what did not, and why you changed direction. Return to earlier steps as needed. Develop the idea into a paper when the question, contribution, and feasible design come together.
Download the skill and begin with a research interest, for example:
I am interested in industrial organization and health economics. Help me find one empirical question with a clear economic contribution, credible identification, and usable public data. Keep a record of the alternatives and why each was retained or set aside.
The downloadable skill provides the general workflow and tools. The ETC map documents how I used it in my own research.