Research on Dynamic Portfolio Rebalancing Strategies with Explicit Consideration of Transaction Costs

Abstract

Classical portfolio theory frequently assumes frictionless markets, but in reality, transaction costs, like fees and market impact, can erode returns and cause excessive turnover. Incorporating these costs transforms rebalancing from a mechanical rule into a strategic decision: determining exactly when and how much to trade to restore desirable exposures efficiently. This study proposes a dynamic rebalancing framework that explicitly models proportional transaction costs within a multi-period optimization setting. By introducing a transaction cost function, portfolio adjustment becomes a rigorous optimization problem balancing expected returns, risk exposure, and total rebalancing costs. This cost-aware framework empowers investors to avoid unnecessary trading, preserving risk control while improving net performance. To solve this, specific algorithmic procedures are designed to enhance computational efficiency in realistic, multi-asset scenarios. Empirical evaluations using historical financial data compare this cost-aware strategy against conventional periodic and threshold-based methods. Performance is assessed across metrics including cumulative return, portfolio volatility, turnover rate, and net returns after costs. The empirical results demonstrate that explicitly integrating transaction costs into optimization significantly improves strategy performance. The proposed model successfully reduces unnecessary trading activity and lowers portfolio turnover while maintaining competitive risk-adjusted returns. Furthermore, sensitivity analyses reveal that transaction cost levels dictate the optimal rebalancing frequency and adjustment magnitude. Overall, this study provides a systematic modeling framework and strong empirical evidence for cost-aware dynamic portfolio rebalancing, offering practical insights for investors navigating complex environments.