Phd student at LSE.
Dates of visit: 01/10/2012 – 05/07/2013
Project Description: Uncertainty is a part of science. Often decisions – important, costly decisions – must be made on the basis of incomplete or uncertain evidence. For example, decisions regarding our response to climate change, economic decisions that rely on models of future economic output, decisions regarding the adoption of new technologies, and so on. I claim that there is a disconnect between the multidimensional, complex nature of scientific uncertainty and the simple models of decision making under uncertainty in decision theory. This project aims at exploring how to model the various kinds of uncertainty that arise in science. Standard probability theory is a reasonable tool for accommodating some, but not all sources of uncertainty in science. I will clarify when probability theory is appropriate and when it is not. I will outline some alternative, more permissive formal methods that allow us to formally model more kinds of uncertainty. I will use climate science as a case study, but my conclusions about scientific uncertainty and decision making should apply more generally.