On 6th October 2016 the RSS Glasgow local group hosted a joint event with the Environmental Statistics section of the RSS and the University of Glasgow’s Boyd Orr Centre for Population and Ecosystem Health on Statistical Ecology.
The audience of 75+ were treated to an excellent double bill event from Professor Ruth King, University of Edinburgh and Dr Diana Cole, University of Kent with the event introduced by Dr Claire Miller from the RSS Glasgow local group and chaired by Prof Jason Matthiopoulos and Dr Duncan Lee representing the Boyd Orr Centre and the Environmental Statistics section respectively.
Professor Ruth King presented work on recent advances in the analysis of multi-state capture-recapture data with applications for wildlife populations. Prof King used illustrative ecology examples to explain that individuals may be recorded as being in different discrete states when they are observed and this may relate to location, disease status, breeding status, etc. For example, the hunger of a snake is related to time since last having eaten as opposed to whether or not the snake has eaten at the previously recorded state. Prof King illustrated how the, often fairly complex, capture-recapture models used for analysing such data can be expressed in the form of a general (partially observed) hidden Markov model permitting a generalized framework for model fitting.
Dr Diana Cole discussed Parameter Redundancy and Identifiability in Ecological Models explaining that to be able to fit or examine a parametric model successfully all the parameters need to be identifiable. Dr Cole provided ecological examples of situations where the parameters are non-identifiable resulting in a model which is parameter redundant and that can be rewritten in terms of a smaller set of parameters. However, she explained that parameter redundancy is not always obvious. Dr Cole presented a new approach for such ecological models based on a hybrid-symbolic numeric method, which is extremely accurate compared to other numeric methods, and is in the process of being automated for application by non-experts.
Tweets and pictures from the event are available at: #RSSGlaEco, https://twitter.com/hashtag/RSSGlaEco