Hello, I am R. J. Bowater and welcome to my Modern Statistical Theory website.
I work as a Research Fellow in the Peninsula Technology Assessment Group (PenTAG) in Exeter University, UK. I have lived for long periods of time in England, Italy and Mexico. Please feel free to contact me via email or my X.com (Twitter) account.
This theory consists of approaches to inference that are empirical (e.g. Neyman-Pearson or Fisherian inference), axiomatic (e.g. orthodox Bayesian inference or imprecise probability theory) or both (despite the considerable difficulties involved in attempting to combine these two usually opposing approaches to inference). Classical statistical theory was always destined to be a failure, but the theory had too much riding on for it to be allowed to fail quickly. Most research work in statistics is still focused on developing aspects of classical statistical theory.
This theory is effectively the analogical approach to inference (i.e. methods of inference constructed and justified via making statistical analogies). It has the feel of a next or final generation statistical theory as not only does it constitute a satisfactory completely general framework for performing statistical inference, but it provides a laboratory for carefully analysing what went wrong and, to some extent, what went right with classical statistical theory.
This website, currently under construction, is dedicated to my work in developing modern statistical theory. A timeline of how I was led to the way of thinking that underlies this theory with references to my work can be found here, while talks that I have given related to this theory can be found here. If you are interested in collaborating with me on this project and thereby in becoming the second person in the world actively working in this research area, then let me know!