Addressing statistical challenges of modern technological advances

ICMS, Bayes Centre, 47 Potterrow, Edinburgh EH8 9BT

24 - 28 June 2019

Organisers



Steve Buckland, University of St Andrews

Ruth King, University of Edinburgh

Rachel McCrea, University of Kent

Byron Morgan, University of Kent

Ken Newman, BioSS and University of Edinburgh


Technological advances in collecting ecological data have expanded rapidly in the last few years, and at a significantly faster rate than the associated and necessary statistical tools for analysing the data to obtain inference on the ecosystems under study. The availability of cheap technological devices suitable to collect the data has led to an explosion of new forms of data; quality of data (e.g. finer temporal/spatial resolution); and quantity of data. Each of these aspects provide new statistical challenges. Specific examples of the new forms or increased quality/quantity of data available include (but are not limited to): advanced geolocation tags; motion/acoustic detector trap arrays; and citizen science data. These lead to a mathematical demand for new statistical methodology that can take full account of the strengths and limitations of the data generated. Challenges that arise with these forms of data that need to be addressed include, for example, developing new finer resolution models (and continuous time/space models); accounting for interactions at both the individual and species level; dealing with computationally intensive likelihoods (and/or big data); dealing with different forms of missing (correlated) data; and efficient model-fitting tools. In addition to the individual challenges raised by the new data available, it is common for multiple different types of datasets be collected on the same ecosystem, leading to further challenges in integrating such data from different sources (of differing quality). However, the potential pay-off for being able to combine different forms of data, often of varying quality and quantity, has huge, and as yet largely untapped, potential.

The workshop programme focuses on these emerging forms of data and the new statistical models and associated model-fitting tools for analysing such data. This will include identifying the key issues that need to be addressed and discussing new potential approaches. The workshop will also provide a platform for identifying new avenues related to the main statistical challenges for modern statistical ecology, build new collaborations and provide the opportunity to explore new interdisciplinary areas.

There will be a Public Lecture on Tuesday 25th June given by Prof. Stephen Buckland, St Andrews University.

Arrangements

Speakers invitations to this workshop were issued by ICMS in February 2019.

The public application is now closed. Decisions will be made on applications by Wednesday 15th May, and applicants will be informed of the decison by email.

ICMS does NOT use third parties to arrange accommodation.  If you are approached by a third party (e.g. Business Travel Management) asking for booking details, please ignore.  If you have any concerns please contact ICMS.

Programme

To be announced

Invited Speakers

Please contact diane.horberry at icms dot org dot uk, or one of the organisers for further details on the invited speakers.