I am a postdoctoral researcher in Statistical Ecology (or ecological statistics) at the University of Bayreuth, Germany. My research focuses on the development of multivariate statistical models for ecological communities. Much of my work revolves around translating modern statistical methodology into tools that ecologists can use, including software development, computational methods, and improving the alignment of statistical models and ecological processes.

A central theme in my research is dimension reduction for multivariate ecological data, which in ecology is referred to as ordination. I have a particular interest in ordination methods, although not so much in the classical techniques that have traditionally dominated the field.

More broadly, I think the analytical tools used in community ecology are often poorly aligned with the ecological processes being studied. While statistical methodology has advanced rapidly, classical multivariate techniques, non-parametric estimators of diversity, and ad-hoc analytical workflows are still widely used. One of the main motivations behind my work, is to help bring everyday analyses closer to modern statistical modeling.

In my view, multivariate mixed-effects models, and particularly Generalised Linear Latent Variable Models (GLLVMs), provide an encompassing framework for community ecology, because of their ability to more closely reflect ecological processes, while still facilitating analysis of everyday community ecological datasets. Much of my research therefore focuses on model-based ordination, including methodological development, practical applications, and improvements in usability, often in connection with the gllvm R-package.

While ordination is an important analytical tool for sparse data, my interests increasingly extend to biodiversity estimation and forecasting, particularly how these statistical model can be used to better understand and predict biodiversity dynamics.

Ultimately, I am interested in contributing to a gradual shift toward model-based approached in community ecology, spanning ordination, diversity estimation, clustering, and related analytical techniques.

Research interests

My professional interests include:

• Model-based ordination using Generalised Linear Latent Variable Models
• Joint Species Distribution Models, insofar they overlap with model-based ordination
• Approximate likelihood methods, especially Variational Approximations
• Computational statistics and scalable inference - models must run before I’ve brewed my coffee (Life is too short for MCMC)
• Numerical optimisation for statistical models (Only a little; it’s difficult to avoid)
• Best practice-statistical analysis, ensuring assumptions are realistic and coherent
• Statistical methods for everyday ecological datasets
• Community ecology

Publications

Please have a look at my google scholar profile.