dr. I. (Igor) Pouchkarev

Conventional performance evaluation methods strongly differentiate between the universe which is used for portfolio construction, and the universe which is used for the performance evaluation. Whilst by composing the portfolio we consider the complete opportunity set, in the last case we use a very restricted, general representation of this opportunity set: a peer group or (a) benchmark portfolio(s). In this thesis we present a conceptual framework, which allows to incorporate the decision-making context of any constrained investment portfolio into the performance evaluation process. The main feature that distinguishes our methodology from conventional performance evaluation methods is that it tackles the performance at the decision-making level: the portfolio weights. We consider all possible portfolios that can be constructed given the specific investment objective(s) as well as the prescribed investment constraints, and then evaluate all these portfolios according to (a) selected performance measure(s). The performance of the investment portfolio is calculated simultaneously and then evaluated against the performance of this complete opportunity set. Consequently, our methodology extends the conventional performance metrics with the insights into the performance of all opportunities that existed at the time of the investment decision.