Graphs as Models of Interventions
from Concrete Causation Hosted by Roland Poellinger
In this talk Roland Poellinger (Munich) gives an outline of Judea Pearl's deterministic approach towards causation (workshop "Concrete Causation", 9 July, 2010). The title of the talk is taken from the programmatic section 2.2 of Pearl's paper "Causal Diagrams for Empirical Research" (Biometrika, Vol. 82, No. 4, 669-709, 1995) which is briefly sketched and commented on as an introduction to Pearl's interventionist account of causal analysis. Further topics: The problems of simple causal networks, interventions as variables, Humphreys' paradox, and causal decision making.
Embed this episode
No transcript yet.