Submitted
Structure-Preserving Randomization Inference
Testing placement effects while holding event structure and the exogenous path fixed.
- Status
- Submitted
- Version
- Method article
- Date
- 2026
- Authors
- Aryan Patel
Abstract
A method-focused project that preserves the observed structure of a decision sequence and the exogenous path, then randomizes placement under a declared null. The design separates an attractive realized outcome from evidence that placement itself carried information.
Research question
Does observed event placement perform differently from placements generated under a structure-preserving null?
Methods
- Structure-preserving randomization
- Declared placement mechanism
- Fixed exogenous path
- Sensitivity to outcome measures
- Reproducibility checks
Data
Public market data used by the project; exact coverage is documented in the manuscript materials.
Results
Submitted method article. An earlier submission was desk-rejected at Risks. The work is not published.
Limitations
Inference depends on the declared placement mechanism, outcome measures, and fixed path. It is not a universal test of timing skill.
- Code availability
- Research materials maintained with the manuscript package.
- Data availability
- Public market data.
- AI disclosure
- AI-assisted software supported literature retrieval, code generation, analysis, critique, and drafting. Deterministic systems produced and checked estimates. A human researcher retained responsibility for the question, design, interpretation, and release.
- Reproduction status
- Reproducibility materials prepared; release status follows the manuscript.