Friday, March 28, 2025
Lecture room B6
Institute for Exact Sciences
Sidlerstr. 5
CH-3012 Bern
16:50-17:40 h
Testing Temporal Causality
The starting point of the talk is a causal model with an explicit temporal structure. That is,
a model capable of describing how a dynamic system evolves over time in an observational as well as
in interventional scenarios. Examples include models of chemical reactions, industrial processes,
economic and financial systems, infectious disease spreading and survival and event processes.
In the first part of the talk I will review a range of different sampling schemes of data
from a dynamic system, which oftentimes leave aspects of the system unobserved. An overarching
goal is nevertheless to characterize all testable constraints on the observed data that are
entailed by the causal model, which is ambitious and very difficult in general. In the second part
of the talk I will focus on two specific schemes: time series data with unobserved coordinates and
steady-state cross-sectional data. In both cases I will present our main results in terms of
deriving testable constraints and developing tests of these constraints.
