A dynamic conditional approach to forecasting portfolio weights

From the autoregressive representation of the portfolio-variance optimization problem, we derive a novel model for conditional portfolio weights as a linear function of past conditional and realized (and, hence, observable) terms. This dynamic conditional weights (DCW) approach is benchmarked against popular model-based and model-free specifications in terms of weights forecasts and portfolio allocations. Next to portfolio turnover and variance, we introduce the break-even transaction cost as an additional measure that identifies the range of transaction costs for which one allocation is preferred to another. By comparing minimum-variance portfolios built on the components of the Dow Jones 30 Index, the proposed DCW attains the best allocations overall with respect to the measures considered, for any degree of risk aversion, transaction costs, and exposure. (C) 2020 International Institute of Forecasters. Published by Elsevier B.V. All rights reserved.

Automated summary

The novel model based on the autoregressive representation of the portfolio-variance optimization problem introduces dynamic conditional weights (DCW) as a linear function of past conditional and realized terms. The DCW approach outperforms popular model-based and model-free specifications in terms of weights forecasts and portfolio allocations, achieving the best allocations overall for different levels of risk aversion, transaction costs, and exposure.