Who gains and who loses on stock markets? Risk preferences and timing matter

This paper uses an agent‐based multi‐asset model to examine the effect of risk preferences and optimal rebalancing frequency on performance measures while tracking profit and risk‐adjusted return. We focus on the evolution of portfolios managed by heterogeneous mean‐variance optimizers with a quadratic utility function under different market conditions. We show that patient and risk‐averse agents are able to outperform aggressive risk‐takers in the long‐run. Our findings also suggest that the trading frequency determined by the optimal tolerance for the deviation from portfolio targets should be derived from a tradeoff between rebalancing benefits and rebalancing costs. In a relatively calm market, the absolute range of 6% to 8% and the complete‐way back rebalancing technique outperforms others. During particular turbulent periods, however, none of the existing rebalancing techniques improves tax‐adjusted profits and risk‐adjusted returns simultaneously.