Dr Sara Taskinen
University of Jyväskylä
In many ecological studies, counts or biomass of interacting species are collected from several sites. Such data are often very sparse, high-dimensional and include highly correlated responses, and the main aim of the statistical analysis is to understand relationships among such multiple, correlated responses. In this talk we show how generalized linear latent variable models can be used to analyse data common in ecological studies. By extending the standard generalized linear modelling framework to include latent variables, we can account for any covariation between species not accounted for by the predictors, species interactions and correlations driven by missing covariates. Fast and efficient maximum likelihood based algorithms for fitting the models will be discussed and simulations are used to study the finite-sample properties of the resulting estimates. It is shown that especially the variational approximation method performs well when fitting GLLVMs. We will also illustrate how GLLVMs can be used when developing new tools for ecological monitoring of Finnish peatlands.
Biography of Sara Taskinen
Dr. Sara Taskinen is an University Lecturer at the Department of Mathematics and Statistics at the University of Jyväskylä. Her main research interests include nonparametric and robust multivariate methods, statistical signal processing and statistical methods in biology and ecology.
Date: Wednesday, 12 July 2017
6:00pm – 6:30pm: Refreshments
6:30pm – 7:30pm: Lecture
7:45pm onwards: Dinner (at a nearby restaurant)
Venue: Room 430, Level 5, Building 4, University of Technology Sydney (this room is also known as the Maths Grid Room CB04.05.430).
Building 4 is located on the corner of Thomas and Harris Street, Ultimo. The best entrance to use is on the Harris Street side, the one immediately outside the ActiveFitness gym. For more details regarding directions, click here.
|Time:||6:00 pm - 7:30 pm|
|Location:||UTS Maths Grid Room CB04.05.430,
Room 430, Level 5, Building 4, University of Technology,