“The Verisimilitude Framework for Inductive Inference” by Olav Vassend

Abstract:
The “likelihood” of a hypothesis given a piece of evidence is the probability that the hypothesis assigns to the evidence. Both Bayesians and likelihoodists use likelihoods to quantify evidential impact, and likelihoods play an important role in frequentist inference as well. However, I show that the likelihood is not always an appropriate way of measuring evidential impact. I then argue in favor of a “verisimilitude framework” for inductive inference, and I give several examples of verisimilitude-based inference procedures that make use of evidential measures other than the likelihood, including inference procedures appropriate for parsimony evaluations of scientific theories and for phylogenetic inference. Finally, I contrast my proposal with a similar recent proposal grounded in decision theory.

Date: 9 November 2017
Time: 3pm to 5pm
Venue: Philosophy Meeting Room (AS3-05-23)

About the Speaker:
Olav Vassend is an assistant professor of philosophy at Nanyang Technological University. He received his PhD in philosophy from the University of Wisconsin—Madison in the spring of 2017. Most of his work is in philosophy of science and formal epistemology, and he is particularly interested in the foundations of statistical inference and inductive inference more generally.

All are welcome