This repository contains the full source code, MCMC analysis, and results for the Phase-transition Linear Model (PLM-FP). This alternative cosmological framework proposes that the observed cosmic acceleration is an emergent effect of a dynamic physical time, driven by the process of large-scale structure formation.
The main scientific results, methodology, and theoretical background are presented in the paper, which is available in the /communications directory and on Figshare:
- View Paper: communications/FigShare/Out/PLM_FP_Figshare.pdf
- Published Version (Figshare): https://figshare.com/articles/thesis/A_Cosmological_Model...
Our analysis demonstrates that the PLM-FP model provides a statistically superior fit to a combination of Supernovae (Pantheon+), Baryon Acoustic Oscillation (BAO), and Cosmic Microwave Background (CMB) data when compared to the standard ΛCDM model.
| Criterion | PLM-FP (7 params) | ΛCDM (6 params) |
|---|---|---|
| χ² | 676,298 | 7,814,451 |
| BIC | 676,350 | 7,814,496 |
| ΔBIC | \multicolumn{2}{c | }{-7,138,146} |
This constitutes decisive statistical evidence in favor of the PLM-FP model.
The model's superior fit and key physical components are illustrated in the figures below. All result plots are located in the /mcmc_analysis/results/ directory.
Figure 1: Hubble Diagram & Residuals
(This visually demonstrates the superior fit to SN Ia data)
Figure 2: Model Parameter Posteriors (Corner Plot)
(Shows the best-fit parameter space found by the MCMC analysis)
Figure 3: Physical Components of the PLM-FP Model
(Illustrates the underlying physics: evolving H(z), time rate, etc.)
- /communications: Contains the final scientific paper.
- /mcmc_analysis: The main source code for the project.
-
- /models: Python implementation of the PLM-FP and ΛCDM models.
-
- /likelihoods: Likelihood functions for SN, BAO, and CMB data.
-
- /runners: Main scripts for executing simulations and generating plots.
-
- /results: Output directory for all MCMC chains (.h5), plots (.png), and data files.
-
- /data: Observational data files.
- Prerequisites: Python 3.10+,
numpy,scipy,matplotlib,emcee,corner,astropy. - Download Data: Ensure the necessary data files (e.g.,
pantheon_plus_data.txt) are present inmcmc_analysis/data/. - Run MCMC Simulation: The main MCMC run can be executed from the root directory via:
python mcmc_analysis/runners/run_mcmc.py --model PLM
- Analyze and Plot: After the simulation completes, generate the comparison plots and statistics:
python mcmc_analysis/runners/compare_models.py python mcmc_analysis/runners/create_publication_plots.py
Milen Krumov
- Independent Researcher
- ORCID: 0009-0008-3957-9060
- Email: krumov.milen@gmail.com