Optimal dynamic portfolio selection with earnings-at-risk

In this paper we investigate a continuous-time portfolio selection problem. Instead of using the classical variance as usual, we use eamings-at-risk (EaR) of terminal wealth as a measure of risk. In the settings of Black–Scholes type financial markets and constantly-rebalanced portfolio (CRP) investment strategies, we obtain closed-form expressions for the best CRP investment strategy and the efficient frontier of the mean-EaR problem, and compare our mean-EaR analysis to the classical mean-variance analysis and to the mean-CaR (capital-at-risk) analysis. We also examine some economic implications arising from using the mean-EaR model.