OxWaSP Seminar: R Ryder (Université Paris-Dauphine) & J Rougier (Bristol)
Speaker 1: J. Rougier (Bristol)
Modelling the eruption processes of a large number of similar but not identical volcanoes
Extreme value theory suggests a three-parameter model for large eruptions for a specific volcano. Unfortunately, we only have reliable records going back a few hundred years at most, and in that time sometimes have only one or two eruptions per volcano. Pooling the volcanoes is unattractive (although it is commonly done); one alternative is to model them exchangeably. I'll discuss an exchangeable model for many volcanoes, and some of the issues that arise in implementing it. These include squeezing information out of unreliable data, transforming the parameters to apply informative priors, and computational approximations based on discretisation (no MCMC!). I will present return period curves for the eleven currently active Japanese stratovolcanoes. This is joint work with Profs Steve Sparks and Kathy Cashman.
Speaker 2: R. Ryder (Université Paris-Dauphine)
Bayesian methods for Historical Linguistics
Languages change through time in a manner comparable to biological evolution. Models have been developed for many aspects of human languages, including vocabulary, syntax and phonology. The complexity of these models, as well as the nature of the questions of interest, make the Bayesian framework quite natural in this setting, which explains why much of the research in Statistics applied to Historical Linguistics uses Bayesian methods. I shall present an overview of various models, starting with Morris Swadesh's failed attempts at glottochronology in the 1950s, then looking at some models developed in the last decade. I shall go into more detail for a model of so-called "core" lexical data by a stochastic process on a phylogenetic tree, with an initial focus on the Indo-European family of languages and on the issue of dating the most recent common ancestor to these languages. This will allow me to discuss issues of model robustness and validation, before I present applications to several data sets. If time allows, I shall conclude with some ongoing work about long-term trends in changes of syntactical features, especially word order. The work on lexical data is joint with Geoff Nicholls; the work on word order is joint with Isabelle Charnavel, Hilda Koopman and Dominique Sportiche.