Probabilistic forecasting guides dynamic decisions

The assets we invest in---whether projects, jobs, or relationships---often improve over time at varying rates. This uncertainty creates a challenge: How do we decide what asset to invest in and when to divest from an underperforming asset? In this paper, we propose a normative model of asset selection and switching. We hypothesize that people use Bayesian function learning to extrapolate asset performance trajectories and choose assets based on anticipated final performances at a task-specific time horizon. We test the model in three experiments (N = 380) that manipulate the setup (selection versus switching), time horizons, and statistical properties of asset performance trajectories. The model outperforms myopic alternatives inspired by foraging and study time allocation. Our work highlights the role of probabilistic forecasting---integrating prior knowledge with observed asset performance---in guiding complex decisions about uncertain future outcomes.