Technical Note—A Robust Perspective on Transaction Costs in Portfolio Optimization

Abstract

We prove that the portfolio problem with transaction costs is equivalent to three different problems designed to alleviate the impact of estimation error: a robust portfolio optimization problem, a regularized regression problem, and a Bayesian portfolio problem. Motivated by these results, we propose a data-driven approach to portfolio optimization that tackles transaction costs and estimation error simultaneously by treating the transaction costs as a regularization term to be calibrated. Our empirical results demonstrate that the data-driven portfolios perform favorably because they strike an optimal trade-off between rebalancing the portfolio to capture the information in recent historical return data, and avoiding the large transaction costs and impact of estimation error associated with excessive trading.

How much should portfolios shrink?

“…As penalizing the deviation from the current portfolio is similar to accounting for transaction costs, this finding is contrary to the earlier findings that accounting for transaction costs in portfolio optimization enhances portfolio robustness and performance: for example, Gârleanu and Pedersen (); DeMiguel et al. (); Olivares‐Nadal and DeMiguel ().…”

“…The above finding conveys an important message as the deviation penalty models with resemble the models that take transaction costs into account (e.g., Gârleanu & Pedersen, ; DeMiguel et al., ; Olivares‐Nadal & DeMiguel, ). Shrinking toward the current portfolio does improve portfolio performance but is less effective than shrinking toward the equal‐weight portfolio regardless of transaction costs.…”

Fuzzy portfolio optimization with tax, transaction cost and investment amount: a developing country case

“…However, transaction costs are significant for cases when portfolio return is low and also for investors who make frequent purchases/sales [15]. Moreover, in recent studies like [25], the importance of transaction cost and investment amount on the portfolio has been comparatively examined by using different methodologies like robust PO and Bayesian PO. According to the authors, investment amount and cost interacts with each other.…”

“…For example, small trades do not impact the market price, and the transaction cost is assumed to be proportional to the amount traded. Larger trades impact the market price, and it is assumed that it results in quadratic transaction cost [24,25].…”

“…Omitting them or deducting the cost from the portfolio return may result in inefficient portfolios. In other words, costs can erode the gains from a trading strategy [25]. Hence, fuzzified Konno-Yamazaki models were extended with the inclusion of tax and transaction costs for different investment amounts in this study.…”