Dynamic Portfolio Choice with Predictable Returns and Transaction Costs for More Frequent Rebalancing

We consider a dynamic portfolio optimization problem when expected returns follow a linear factor model and transaction costs are quadratic. A closed-form solution is derived for the optimal portfolio policy, which is a gradual trading towards a moving aim portfolio. To pursue the short-horizon investment opportunities, the optimal portfolio policy requires frequent rebalancing with trade-off between expected returns, transaction costs and estimation errors. We introduce two powerful short-horizon factors to predict the subsequent daily returns. We implement the optimal trading strategy for 18 international assets. The dynamic portfolio strategy is tested both in-sample and out-of-sample. For both tests, the optimal dynamic strategy achieves superior net Sharp ratio and substantial cumulative profits relative to more naive benchmarks