Evaluating a Factor: the Information Coefficient

Before you ever run a backtest, you can measure whether a signal has any predictive power with the Information Coefficient (IC) — the rank correlation between today’s signal and tomorrow’s return across the cross-section. We build a real signal and a fake one and learn to tell them apart.

Key takeaways

1. 1.IC is the rank correlation of a signal vs. next-period return — a positive, persistent mean IC is a real edge.
2. 2.Judge the IC information ratio, not a single day; real equity ICs are small (≈ 0.02–0.05).
3. 3.IC decay sets your holding horizon and turnover.
4. 4.A noise signal has ≈ zero IC — the same lesson as p-hacking, seen cross-sectionally.