Why Low-Sharpe Signals Can Improve a Diversified Portfolio

Summary

The article uses simulations to argue that a strategy with a modest standalone Sharpe ratio may still add value when its returns have low correlation with an existing portfolio. It models twenty daily signals with an annualized Sharpe ratio of 0.6, varies their pairwise correlation, and repeats the experiment across many simulated samples. While adding signals does not necessarily raise expected annual return, the article reports that lower correlation narrows the return distribution and can substantially improve the portfolio’s risk-adjusted performance.

It also examines estimation uncertainty. Shorter observation periods leave Sharpe estimates more exposed to chance, and the article reports that many individual signals in its setup fail a conventional significance test even though combinations of sufficiently uncorrelated signals perform better. These findings illustrate diversification and the value of evaluating signals in portfolio context. They depend on the simulation’s assumptions, including ignored transaction costs and specified signal behavior; they do not show that a real low-Sharpe strategy will be profitable or that correlations will remain stable.

Key ideas

• A signal’s contribution depends partly on how its returns correlate with the existing portfolio.
• Combining low-correlation signals can raise portfolio Sharpe even when each signal has a modest standalone Sharpe.
• The simulations show greater uncertainty in estimated performance when the observation period is shorter.
• Many individual signals may appear statistically insignificant while a diversified combination is more significant.
• The results omit transaction costs and rely on simulated assumptions, so they do not guarantee real-world performance.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.