How Decision Frequency, Skill, Costs, and Volatility Shape Long-Term Returns

Summary

This article examines how often an investor should make decisions when each decision reflects some degree of skill and incurs trading costs. It models repeated independent outcomes and argues that expected compounded return depends nonlinearly on skill, market volatility, transaction costs, and decision frequency. Too few decisions can leave potential opportunities unused, while too many can let costs overwhelm the edge; both extremes may require greater skill to produce a positive result.

Its empirical analysis uses US equity and government bond futures data from 2008 to 2018, with closing bid–ask spreads as a cost proxy and Monte Carlo simulations across decision frequencies. Under a constant-skill assumption, the estimated return-maximizing frequencies differ substantially between the two markets. The article also illustrates that optimal frequency may change when skill varies with frequency and that losses can outweigh comparable gains. These conclusions depend on simplified assumptions, historical estimates, and static cost and volatility measures, so the reported frequencies are not universal prescriptions.

Key ideas

• Expected compounded return depends jointly and nonlinearly on skill, trading costs, volatility, and decision frequency.
• Both very low and very high decision frequencies can demand greater skill to overcome foregone opportunities or transaction costs.
• Monte Carlo estimates using US equity and government bond futures show different return-maximizing frequencies under a constant-skill assumption.
• The frequency that maximizes expected return need not be the frequency at which an investor’s skill is greatest.
• Asymmetric compounded outcomes make overestimating skill potentially more damaging than underestimating it.

Tags

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