Sean Koth

Finance student at Fordham University's Gabelli School of Business. Focused on quantitative methods in biotech investing — specifically the information asymmetries that exist around FDA decision dates.

ApprovalAlpha started as a backtest project: can publicly available data — clinical trial registries, SEC filings, FDA correspondence — actually predict approval outcomes better than chance? The answer turned out to be yes, at 0.70 AUC on a held-out test set. The engine now runs live against any public biotech ticker.

The goal is a research tool that surfaces signal most analysts skip — enrollment pace relative to protocol targets, insider buying patterns ahead of PDUFA dates, CRL history that compounds risk across programs — and puts it into a calibrated probability estimate rather than a qualitative rating.