The Greenium: What a Green Bond Saves a Government in Yield
Germany issues green and conventional federal bonds that are identical except for the green label. The yield gap between the twins is the best measure of the greenium there is.
Quantitative research in economics and finance. Projects span causal inference, urban economics, labor markets, financial risk, public economics, investment, energy policy, and data engineering. Each project page includes methodology, findings, and interactive tools where applicable.
Germany issues green and conventional federal bonds that are identical except for the green label. The yield gap between the twins is the best measure of the greenium there is.
The largest IPO in history, read as an investment, a risk register, and a market event, then used to make the engineering case for a Social Wealth Fund funded by an equity condition on public support. A living analysis, scored weekly.
Causal analysis of ridership, fare equity, and transit strain across six 2026 World Cup host cities. Synthetic control, panel DiD, event studies, and an interactive D3/Leaflet dashboard. Living analysis updated weekly.
Calibrated real-time recession probabilities built only from data available at each historical date, with uncertainty bands and an honest out-of-sample horse-race against the term-spread probit and the Sahm rule. Updated weekly.
A calibrated probability-of-default model on Freddie Mac single-family loans, treating calibration, cost-sensitive thresholds, and SHAP reason codes as the headline, not an afterthought.
A staggered difference-in-differences study of ACA Medicaid expansion on coverage, access, and health.
How much of Canada's 1 Apr 2025 carbon-charge removal (~17.6 ¢/L) reached the pump? Event study controlling for crude confounding, Quebec difference-in-differences, rockets-and-feathers ECM, and an independent 2026 excise replication.
TWFE DiD on 48.5M BIXI trip records estimates two REM openings' effect on bike-share demand. Matched estimates: +10.4% and +2.7%. Nearest-neighbor matching revealed 80% of the naive Deux-Montagnes effect was confounding bias from dissimilar controls.
Hedonic OLS on 8,778 Montreal Airbnb listings shows the naive transit premium reverses sign once centrality enters: rail proximity is a centrality effect in disguise, and conditional on location, rail adjacency carries a slight price discount.
Causal inference pipeline (DoWhy, Pearl's do-calculus) recovers a backdoor-adjusted price elasticity of +0.119 for bitcoin mining energy demand; 106-country monthly panel confirms cross-sectional heterogeneity.
Hedonic wage regression on ~8,000 job postings across the UK, Canada, and Australia estimates wage premia associated with AI-adjacent skills in the post-GPT labor market.
A production-quality risk engine computing Value-at-Risk three ways (historical, parametric, Monte Carlo) with GARCH(1,1) volatility modeling, formal backtests, and an interactive Streamlit dashboard.