This paper resolves a question proposed in Kardaras and Robertson (2011): how
to invest in a robust growth-optimal way in a market where precise knowledge of
the covariance structure of the underlying process is unavailable. Among an
appropriate class of admissible covariance structures, we characterize the
optimal trading strategy in terms of a generalized version of a principal
half-eigenvalue of a Pucci extremal operator and its associated eigenfunction.
We consider a zero-sum stochastic differential controller-and-stopper game in
which the state process is a controlled jump-diffusion evolving in a
multi-dimensional Euclidean space. In this game, the controller affects both
the drift and the volatility terms of the state process. Under appropriate
conditions, we show that the lower value function of this game is a viscosity
solution to an obstacle problem for a Hamilton-Jacobi-Bellman equation, by
generalizing the weak dynamic programming principles in [3].
Our goal is to resolve a problem proposed by Karatzas and Fernholz (2008):
Characterizing the minimum amount of initial capital that would guarantee the
investor to beat the market portfolio with a certain probability as a function
of the market configuration and time to maturity. We show that this value
function is the smallest supersolution of a non-linear PDE. As in Karatzas and
Fernholz (2008), we do not assume the existence of an equivalent local
martingale measure but merely the existence of a local martingale deflator.