Supplementary data and analysis code for "Hierarchical Bayesian analysis of the radial acceleration relation scatter: a causal diagram approach to error calibration and intrinsic diversity,"

This archive contains Python scripts and intermediate results for a hierarchical Bayesian analysis of the radial acceleration relation (RAR) scatter in 171 SPARC galaxies. The analysis uses a directed acyclic graph (DAG) to encode causal assumptions, hierarchical Bayesian inference to simultaneously estimate intrinsic inter-galaxy scatter (sigma_halo) and error scale factors (alpha_D, alpha_Inc), and do-calculus to decompose the scatter into causal components. Contents: scripts/ : 9 Python scripts (numbered 00-08) that reproduce all results and figures results/ : 13 CSV files and 2 NPZ files containing MCMC posterior samples and summary statistics src/ : Data loading module for the SPARC database README.md : Detailed documentation of all files, dependencies, and reproduction instructions