Risk management and dynamic portfolio selection with stable Paretian distributions

This paper assesses stable Paretian models in portfolio theory and risk management. We describe an investor's optimal choices under the assumption of non-Gaussian distributed equity returns in the domain of attraction of a stable law. In particular, we examine dynamic portfolio strategies with and without transaction costs in order to compare the forecasting power of discrete-time optimal allocations obtained under different stable Paretian distributional assumptions. We also consider a conditional extension of the stable Paretian approach and compare the model with others that consider different distributional assumptions. Finally, we empirically evaluate the forecasting power of the model for predicting the value at risk of a heavy-tailed return series.