Code and Reproducibility Package for a Preservation-Aware Framework for Nested Environmental Comparisons
Year: 2026
Type: article
Abstract: This software package provides the analysis and simulation code supporting the study "Distinguishing Fine-Scale Dominance from Reference-Relative Concentration: A Preservation-Aware Framework for Nested Environmental Comparisons". The study addresses a fundamental problem in nested environmental comparisons: identifying the level at which disagreement is concentrated does not establish whether fine-scale disagreement is unusually large relative to its parent level. The computational framework connects exact hierarchical decomposition, a log-transformed mean-absolute fine-to-parent statistic, and a conditional unit-membership reference under explicit preservation constraints. The package includes: Exact hierarchical decomposition and empirical comparison procedures. Synthetic experiments evaluating calibration and detection under matched nulls and group-aligned alternatives. Group-size mismatch and temporal-dependence experiments. Spatial-dependence boundary experiments. Leave-one-change-year-out reference analyses. Empirical spatial-autocorrelation diagnostics. Geographically constrained membership-reference sensitivity analyses. Configuration files, execution scripts and documentation. The primary membership reference preserves complete unit time series, the number of groups and the sample-specific group-size vector while reallocating group membership. Secondary geographically constrained references evaluate alternative conditional comparisons. The empirical application examines four spatially resolved CO₂ inventories over 2000–2024 using a fixed national–province–city hierarchy. The package does not redistribute third-party inventory datasets or geographic boundary files. Users must obtain these materials from their original providers and comply with the relevant access conditions and licences. The software is released under the MIT License.
Authors: Mi, Guofang, An, Yihui
Institution: Inner Mongolia University of Finance and Economics
Language: English
Subfield: Economics and Econometrics
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7233888091
