Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | English |
| Published: |
Zenodo
2025
|
| Subjects: | |
| Online Access: | https://doi.org/10.5281/zenodo.17794063 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>This repository provides a reproducible computational framework for studying the dynamics of iterated Pearson-correlation operators on random matrices across a wide range of dimensions. It accompanies the empirical and theoretical results of the associated research project on convergence laws, and fixed-point structure arising from repeated correlation transformations.</p> <p>The framework generates initial matrices with independent entries in the interval [−1,1], applies the centering–normalisation–Gram operator iteratively, and records step sizes, contraction ratios, and convergence statistics. It implements scalable batch execution over thousands of independent trials, together with quantitative diagnostics and figure-production scripts.</p> <p>Main features include high-precision correlation iteration routines, dimension-independent parallel execution, automated logging of trajectory statistics (Δk, Ek, ρk), statistical analysis modules for contraction laws and universality testing, and scripted production of the tables and figures used in the accompanying paper.</p> <p>This framework enables reproducible experimentation on the convergence behaviour of correlation iterations, universality across vast initialization spaces, and empirical fixed-point properties of correlation matrices. It archives all scripts and computation tools required to regenerate the results reported in the paper. Future versions may extend analytical modules, incorporate symbolic verification capabilities, or provide accelerated implementations.</p> <p><strong>Access note:</strong><br>This Zenodo record is currently under <strong>restricted access for peer review</strong>. A private access link has been provided to journal editors and reviewers. The repository will be made <strong>publicly available upon acceptance</strong> of the associated manuscript.</p>