Analysis code supporting "Solar and Wind Expansion and Uneven Fossil-Generation Displacement Across 29 Emerging and Developing Economies"

Résumé fourni par la source

This repository contains the reproducible analysis code for the study “Solar and Wind Expansion and Uneven Fossil-Generation Displacement Across 29 Emerging and Developing Economies.” The software is provided as a Google Colab notebook and a Python script and reproduces the complete analytical workflow for 29 countries over 2010–2024. The code: imports and harmonizes Ember Yearly Electricity Data, World Bank World Development Indicators, and optional Our World in Data energy series; constructs the balanced country-year and annual-change panels; calculates the gross fossil-displacement ratio, clean-growth coverage, electricity-balance residuals, and annual transition categories; estimates two-way fixed-effects models with country-clustered, Driscoll–Kraay, and restricted wild-cluster-bootstrap inference; performs influence, multicollinearity, normalization, and leave-one-country-out robustness checks; evaluates prospective structural, ex-post diagnostic, and full-balance logistic classifiers using leave-one-country-out validation; produces discrimination, calibration, Brier-score, precision-recall, and country-bootstrap uncertainty estimates; exports the processed datasets, out-of-fold predictions, fold-specific coefficients, manuscript tables, supplementary tables, figure source data, and publication figures. The default configuration uses Python 3.12, a fixed random seed of 42, 999 wild-cluster-bootstrap replications, and 2,000 country-cluster bootstrap replications for the principal classification analyses. Required package specifications are supplied in requirements.txt and environment.yml, and execution instructions are provided in README.md. The repository does not redistribute the original public source datasets. Users must obtain the required Ember and World Bank files, and the optional Our World in Data files, from their original providers and upload them using the filenames documented in the README. Licence: MIT License.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.