Simulation code for "A Bernstein–von Mises Theorem for Parametric Competing Risks under Hybrid Censoring"

Description

Python code reproducing all numerical results of the paper "A Bernstein–von Mises Theorem for Parametric Competing Risks under Hybrid Censoring" (Mathematics, MDPI, 2026). The script regenerates, from fixed random seeds, Table 1, Table 3, and Figure 1. No empirical data are used; all results are Monte-Carlo simulations. Requires Python 3.9+, numpy, scipy and matplotlib.

Steps to reproduce

1. Download article_simulation.py, requirements.txt, and README.md. 2. Install dependencies: pip install -r requirements.txt 3. Run the script: python article_simulation.py (a few minutes on a standard laptop). 4. The script prints Table 1 (TV distance / coverage) and Table 3 (Weibull shape coverage) to the console, and writes Figure 1 (posterior vs Gaussian limit) as fig_bvm.pdf in the working directory. All reported numerical results are reproduced exactly from the fixed random seed (SEED = 7).

Institutions

• National University of Uzbekistan

  Tashkent, Tashkent

Categories

Licence