Quant Memo

Information Ratio

Active return per unit of tracking error; measures excess performance vs a benchmark in backtesting.

Definition

Information ratio (IR) = (R_strategy − R_benchmark) / TE, where TE is the tracking error (standard deviation of the strategy’s return minus the benchmark return). It measures active return per unit of active risk.

Why it matters for backtesting

  • Active management: Standard metric for long-only or benchmark-relative strategies.
  • Consistency: High IR implies the strategy consistently beats the benchmark without large deviations.
  • Capacity: Strategies with high IR and low tracking error may be more scalable.

Limitations

  • Only meaningful when a clear benchmark exists.
  • Sensitive to the benchmark; changing benchmark changes IR.
  • Does not account for absolute level of return or drawdown.

Linked concepts

Alpha, beta, Sharpe ratio, tracking error.

ShareTwitterLinkedIn