I'm an economics student at Yokohama National University in Japan. I'm interested in investment research, particularly how Japanese companies earn money and how economic data can inform a research thesis. I use Python to make the data, assumptions, and calculations behind my work easy to inspect.
A Streamlit workspace for researching USS (4732) and MonotaRO (3064). It brings together five annual reporting periods per company, source-linked financial statements, and editable earnings and forward P/E scenarios. Each reported figure can be traced back to an issuer document and page. Project and setup · Valuation methodology
A reproducible U.S. unemployment forecasting study that uses the data available at each historical forecast date. The persistence benchmark had the lowest observed error in the main evaluation, and the uncertainty intervals left the ranking inconclusive. The repository includes the data provenance, forecast-level results, and tests for information leakage. Project and results · Research brief
I keep reading and tooling lists for Japanese equity research, macroeconomics, Python research apps, and GitHub workflows. Browse my Stars lists