On the choice of baselines in multiattribute portfolio analysis: A cautionary note

In multiattribute portfolio optimization, a decision maker must evaluate a number of projects on multiple dimensions and then select the set of projects that optimizes the portfolio's overall value. In this note, we discuss the importance of establishing an appropriate baseline score for not doing a project in multiattribute portfolio analysis. We believe that practitioners often implicitly assume that not doing a project results in the worst possible score on all attributes. We argue that this assumption is often inappropriate and may lead to incorrect recommendations.