Seymour Geisser Distinguished Lecture  October 18, 2007
University of Minnesota
School of Statistics
College
of Liberal Arts

Conditional Independence, Imprecise Probabilities, Null-Events and Graph-Theoretic Models

Teddy Seidenfeld
Department of Statistics
  Carnegie Mellon University

Thursday, October 18, 2007
3:30 PM, Ralph Rapson Hall 31
Minneapolis, East Bank Campus
Social following seminar at 4:30 PM, 300 Ford Hall

 

Abstract

In this presentation I review two recent challenges to the use of conditional probability as an account of probabilistic irrelevance – challenges that threaten both the adequacy of the likelihood principle and that threaten to make invalid the use of familiar graph-theoretic techniques for computing conditional independencies.  

 The challenges emerge in the following two ways. 

1.           The first challenge arises within IP theory – Imprecise Probability theory – where a set of probability distributions is used to depict uncertainty.  That is, in IP theory there are distinct lower and upper probabilities for events, rather than what comes from using a single probability distribution.  The straightforward account of probabilistic irrelevance in IP theory – that conditioning does not change the lower and upper probabilities for an event – is not a symmetric relation! 

2.           The second challenge arises with (real-valued) conditional probability given a null event, i.e. when the conditioning event has 0 probability, as in the theory of full conditional distributions due to work by de Finetti-Dubins-Krauss.  This, too, produces an asymmetric irrelevance relation.   Of course, the second problem can be embedded within IP theory, as when the conditioning event is null for at least one of the probability distributions in the IP set. 

I offer responses to these challenges that return us to a symmetric (“Bayesian”) analysis of probabilistic irrelevance in each case. 

This work on these two challenges reflects collaborations, respectively, (1) with Mark Schervish and Jay Kadane of CMU, and (2) with Fabio Cozman of U. Sao Paulo.