Although the consideration of foreign investments may have a positive impact
on the overall market risk of the portfolio through diversi cation, it also adds
a new source of uncertainty due to changes in the value of the currency. We
investigate portfolio optimization models that account separately for the local
asset returns and the currency returns, providing the investor with a full
investment strategy. We tackle the uncertainty inherent to the estimation of
the parameters with the aid of robust optimization techniques. We show how,
by using appropriate assumptions regarding the formulation of the uncertainty
sets, the original non-linear and non-convex models may be reformulated as
second order cone or as semide nite programs. Additionally to the guarantees
provided by robust optimization, we consider the use of hedging instruments
such as forward contracts and options. The proposed hedging strategies are
implemented from a portfolio perspective, and therefore do not depend on the
individual value or behavior of any particular asset or currency. Hedging decisions
are taken at the same time as investment decisions in a holistic approach
to portfolio management. While dynamic decision making has traditionally
been represented as scenario trees, these may become severely intractable and
di cult to compute with an increasing number of time periods. We present an
alternative approach to multiperiod international portfolio optimization based
on an a ne dependence between the decision variables and the past returns.
We add to our formulation the minimization of the worst case value-at-risk and
show the close relationship with robust optimization. The proposed theoretical
framework is supported by various numerical experiments with simulated and
historical market data demonstrating its potential bene ts.