Statistical Ecology seminar double bill
Join us on the 6th October 2016 from 17.30 - 18.30 for this statistical ecology seminar double bill followed by drinks and nibbles. The event will be jointly hosted with the RSS Environmental Statistics Section and the Boyd Orr Centre for Population and Ecosystem Health
Recent advances in the analysis of multi-state capture-recapture data.
Ruth King, Thomas Bayes' Chair of Statistics at the University of Edinburgh.
Capture-recapture data are commonly collected on wildlife populations. The exact form of the data collection process and modeling assumptions permits different demographic parameters of interest to be estimated. In this talk we will focus on multi-state capture-recapture data, where individuals may be recorded as being in different discrete states when they are observed. For example, this may relate to location, disease status, breeding status etc. We will show how these, often fairly complex, capture-recapture models can be expressed in the form of a general (partially observed) hidden Markov model (HMM). This permits a generalized framework for a range of different models, including likelihood formulation and associated model fitting techniques. Real examples will be used to demonstrate the different models.
Parameter Redundancy and Identifiability in Ecological Models.
Diana Cole, Senior Lecturer in Statistics at the University of Kent.
To be able to fit or examine a parametric model successfully all the parameters need to be identifiable. If the parameters are non-identifiable the model can be rewritten in terms of a smaller set of parameters, and is termed parameter redundant. Parameter redundancy is not always obvious, in which case the definitive method for detecting parameter redundancy involves calculating the rank of a matrix, which is expressed symbolically. Although numeric methods also exist, these can lead to the wrong conclusion. Models used in ecology are becoming more realistic but at the same time more complex. This poses a problem for this symbolic method as computers run out of memory calculating the rank of the appropriate matrix. An extended symbolic method exists but this is mathematically complicated. An alternative solution is a hybrid-symbolic numeric method. This method combines symbolic and numeric methods to create an algorithm that is extremely accurate compared to other numeric methods but is also more straightforward to use. We demonstrate the advantages of this method and show how it can be used in ecological models, such as capture-recapture models.
Twitter: Join the discussion and post questions using the hashtag #RSSGlaEco