Research
Comparing decision frameworks on identical history.
My Honors research asks one question: when different decision frameworks are evaluated on identical historical information, how do they actually perform? Answering it rigorously means building the simulation and evaluation infrastructure to replay decades of history under each framework and score them the same way. This is active work at Oregon State University; the methodology and infrastructure are the contribution.
The contribution
A controlled, reproducible environment where human, quantitative, and AI-driven decision systems act on exactly the same information and are scored on equal footing — so they can be compared rigorously rather than anecdotally.
The experiment
Four frameworks, one history
The same market history, replayed under each framework and scored on equal footing.
- 01
Human decision-making
Discretionary judgment as a baseline.
- 02
Quantitative models
Rules-based, statistical frameworks.
- 03
Human + AI systems
AI augments; the person decides.
- 04
AI-first systems
Model-driven decision processes.
Where the work is
Built, and still to build
The methodology and the infrastructure are the point. Here is what exists versus what is designed but not yet run.
Historical replay engine
Deterministic playback of decades of data.
In progressFramework-agnostic harness
Drop in a decision system and score it under identical conditions.
In progressEvaluation methodology
Defining what “good” means for a decision system, reproducibly.
In progressLarge-scale evaluation runs
Designed for scalable compute; not yet run.
Designed, not yet run
Where it meets the engineering
Some of these ideas already run in real software: in Agentic CRM, AI recommendations stay subject to human approval. The research informs the engineering; the engineering pressure-tests the research.