Kaya Identity Regression

This project decomposes national CO₂ emissions drivers using the Kaya Identity framework, estimating country-level coefficients for population (p), GDP per capita (g/p), energy intensity (e/g), and carbon intensity (f/e). I built a reproducible RMarkdown pipeline that generates a clean country-by-coefficient table and automates multi-country processing with purrr-based functional workflows. To communicate cross-country structure and scale effects, I produced scatter plots where each country is a point and population is encoded by point size, with axes defined by pairwise combinations of Kaya coefficients (e.g., p vs g/p, g/p vs e/g). The visual diagnostics highlight how emissions pathways differ by development level and energy system characteristics, while also surfacing outliers that warrant closer data and model checks. Overall, the work demonstrates an end-to-end workflow for factor decomposition, comparative analysis, and executive-ready visualization.