Portfolio selection with parameter uncertainty under [formula omitted] maxmin mean–variance criterion

Abstract

We consider a mean–variance portfolio selection problem with uncertain model parameters. We formulate the mean–variance problem under the α maxmin criterion, in which the investor has mixed ambiguity aversion and ambiguity seeking attitudes and solves a convex combination of max–min and max–max optimization problems. By the Lagrangian method, we obtain the efficient portfolio and quasi-efficient frontier in closed form. We provide comparative statics of the quasi-efficient frontier to various parameters.