The Royal Statistical Society holds a special place in the history of Bayesian statistics. Britain has produced many of the field's most influential figures—from Thomas Bayes himself to Harold Jeffreys, Dennis Lindley, Adrian Smith, and David Spiegelhalter—and the RSS has served as the institutional home for much of this intellectual tradition. The society's journals, read papers, and discussion meetings have been venues for landmark contributions and historic debates in Bayesian methodology.
Historical Significance
The RSS was founded in 1834 as the Statistical Society of London and received its royal charter in 1887. Throughout its history, the society has been a forum for statistical innovation. Thomas Bayes was a Fellow of the Royal Society (a separate institution), but the intellectual tradition he founded has been nurtured extensively within the RSS community.
The Statistical Society of London is founded, later becoming the Royal Statistical Society.
Dennis Lindley's influential paper "The Use of Prior Probability Distributions in Statistical Inference and Decisions" is discussed at the RSS, marking a turning point in the British Bayesian revival.
Adrian Smith and others present Bayesian work at RSS read-paper meetings, furthering the acceptance of Bayesian methods in British statistics.
RSS journals publish foundational work on MCMC methods, including contributions from the Cambridge and London Bayesian groups.
The RSS tradition of "read papers"—major papers presented at society meetings followed by formal discussion and published with the discussion—has been the setting for some of the most important moments in Bayesian history. The discussion format encourages rigorous critique and has produced exchanges that clarified foundational issues in Bayesian methodology, including debates about prior specification, objective versus subjective Bayes, and the role of MCMC in modern statistics.
Journals
The RSS publishes three series of the Journal of the Royal Statistical Society: Series A (Statistics in Society), Series B (Statistical Methodology), and Series C (Applied Statistics). Series B, in particular, has published many influential Bayesian papers, including Adrian Smith and Gelfand's seminal 1990 paper on the Gibbs sampler that helped launch the MCMC revolution. The society also publishes Significance, a magazine aimed at a broader audience.
The British Bayesian Tradition
Britain has a uniquely rich Bayesian tradition. Harold Jeffreys developed his theory of scientific inference at Cambridge, providing an objective Bayesian framework that remains influential today. Dennis Lindley championed the subjective Bayesian approach and trained a generation of Bayesian statisticians at University College London. Adrian Smith (later Sir Adrian Smith) and his collaborators at the University of Nottingham and Imperial College London developed the computational methods—particularly MCMC—that made modern Bayesian statistics possible. David Spiegelhalter's BUGS software, developed at the MRC Biostatistics Unit in Cambridge, brought Bayesian methods to applied researchers worldwide.
"The RSS has been the stage for some of the most consequential discussions in the history of statistics. Many of these discussions have concerned the Bayesian approach, and they have shaped the field as we know it today."— Adrian Smith