Research preview

The story of a Shang‑Dynasty queen wielding a sword captures the imagination, just as today’s financial data can feel like a battlefield of endless variables. This article connects the intrigue of that narrative with practical lessons from a graduate‑level course on big data and machine learning for financial economics, showing how quantitative traders can turn high‑dimensional information into robust trading signals. The Curse of Too Many Predictors When a model includes dozens or hundreds of predictors, it often fits historical data almost perfectly. In‑sample error shrinks because the model can “explain” random noise, but out‑of‑sample performance deteriorates—a classic case of over‑fitting....