Anthony O’Hagan
School of Mathematics and Statistics, University of Sheffield, UK
Publication details: Submitted to SIAM/ASA Journal of Uncertainty Quantification 20, 2013.
This article is written in the spirit of helping recent efforts to build bridges between the community of researchers in fields such as applied mathematics and engineering, where the term UQ began, and the community of statisticians who work on problems of uncertainty in the predictions of mechanistic models. It is addressed to researchers and practitioners in both communities.
The first purpose of the article is to explain polynomial chaos, one of the key tools of the first community, in terms that will be readily understood by a statistician in the second community. The second purpose is to explain to researchers in the first community some aspects of PC, both in theory and in practice, that a statistician might regard as deficiencies or limitations.