[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Avviso di seminario
Avviso di seminario
On specifying Graphical Models for Causation, and the
Identification Problem
David A. Freedman
Statistics Department
University of California at Berkeley
Giovedì 8 maggio, ore 15.00 e Venderdì 9 maggio, ore 15.00
Sala Riunioni Disa
Facoltà di Economia
Università di Trento
Via Inama 5, 38100 Trento
Abstract
In these talks (which are mainly expository), I will set up graphical
models for causation, having a bit less than the usual complement
of hypothetical counterfactuals. Assuming the invariance of error
distributions maybe essential for causal inference, but the errors
themselves need not be invariant. Graphs can be interpreted using
conditional distributions, so that we can better address
connections between the mathematical framework and causality in
the world. The identification problem is posed in terms of
conditionals. As will be seen, causal relationships cannot be
inferred from a data set by running regressions unless there is
substantial prior knowledge about the mechanism that generated
the data. The idea can be made more precise in several ways. The
paper is availabe on the web, at
http://www.stat.berkeley.edu/~census/601.pdf
Referente: Prof. Klaus Haagen
Per informazioni: 0461 882368