Statistics Colloquium: Analogy-based inference: a universal school of statistical inference
Presented by Jamie Bowater, Research Fellow, Department of Public Health and Sports Sciences, University of Exeter, England
DATE: Friday, October 3, 2025, 10:00 AM. Virtual Presentation
Webex Link: Link
Abstract: Standard statistical theory has arguably proved to be unsuitable as a basis for constructing a satisfactory completely general framework for performing statistical inference. For example, in using empiricist theory, which attempts to justify methods of inference through their empirical performance when used repeatedly, inferences made about a population quantity are only relevant to a reference class of cases, which will inevitably include unobserved cases, rather than being directly relevant to the case at hand. Also, in using axiomatic theory, which is based on the idea of deducing statistical inferences about a population of interest from a given set of universally acceptable axioms, the difficulty is encountered of finding such axioms that are weak enough to be widely acceptable, but strong enough to lead to methods of inference that can be regarded as being efficient. These observations justify the need to take a serious look at an alternative avenue through which statistical inference may be performed, and in particular, the one that is offered by analogy making. In doing this, the use of analogy making will be explored in this talk not just as a supplementary means of understanding how statistical methods work, but as the fundamental basis of an up until now overlooked but nevertheless major school of statistical inference.
Bio: Jamie is currently a Research Fellow in the Department of Public Health and Sports Sciences of the University of Exeter in England. Before arriving in Exeter, he worked as an assistant/associate professor in statistics at Lancaster University in the UK and three universities in Mexico (UAM, UAQ and UDLAP). He has also worked as a researcher in statistical theory/evidence synthesis at the Universities of Warwick, Bath, Birmingham and Bristol in the UK and the University of Padua in Italy. Over the years, Jamie has gained ample experience in statistical consultancy and meta-analysis. He has been interested in the controversies that surround the fundamental nature of statistical inference since his days as a postgraduate student and has maintained this interest throughout his entire research career.