Seminar: 1:00-2:15pm EDT
Student Session: 2:30-3:30 EDT
Hidden Markov Models: The Skeleton in the (Statistical Ecology) Cupboard
Hidden Markov models (or state-space models) are a very efficient and convenient way of representing temporal time series data, where the observed data are dependent on unobserved underlying factors (or states) that may change over time. These models are intuitively and conveniently described via two separate but linked processes: (i) the system, or transition, process that describes the unobserved states; and (ii) the observation process that links the observed data to the (unknown) state of the system process. In this talk we provide a brief introduction to hidden Markov models and demonstrate their applicability to many common models within statistical ecology. We also discuss the usefulness of specifying these models in the hidden Markov model framework.
Ruth King is the Thomas Bayes’ Chair of Statistics in the School of Mathematics at the University of Edinburgh. Her current external roles include President of the International Biometrics Society: British and Irish Region; and Deputy Director of the National Centre for Statistical Ecology. She was elected a Fellow of the Learned Society of Wales in 2017; and the Royal Society of Edinburgh in 2018. Her research interests focus on developing new statistical methodology for analysing data particularly within ecology and epidemiology.
The student session after the talk will allow students to ask Ruth questions about her research, the talk, the recommended paper or career opportunities. If you’re a student, make sure to register for this session.
This month’s paper is Statistical Ecology by Ruth King. The paper and associated supplementary file provide useful background for the talk, describing a series of statistical ecology models and their specification via a hidden Markov model. The paper focusses on concepts of hidden Markov model and not so much on technicalities.
King, Ruth, Statistical Ecology (January 2014). Annual Review of Statistics and Its Application, Vol. 1, Issue 1, pp. 401-426, 2014, Available at SSRN: https://ssrn.com/abstract=2405891 or http://dx.doi.org/10.1146/annurev-statistics-022513-115633