Characterization of time-consistent sets of measures in finite trees

In this paper we give an alternative characterization for time-consistent sets of measures in a discrete setting. For each measure \mathbb{P} in a time-consistent set \mathcal{P} we get a distinct set of predictable processes which in return decribe the \mathbb{P} uniquely. This implies we get a one-to-one correspondence between time-consistent sets of measures and sets of predictable processes with specific features. -- Multiple Priors ; Time-Consistency ; Ambiguity ; Uncertainty Aversion