Mathematics and Statistics

Temporal changes in environmental monitoring data are studied to understand underlying processes of environmental pressures or effects of policy decisions. Often environmental data sets have specific properties that limit the use of traditional statistical methodology, such as temporal and spatial dependence, sparsity of data, outliers and outlying episodes, or values below a reporting limit. 

For further information contact Claudia von Brömssen

Statistical extreme-value analysis is concerned with the study of distributions for extreme observations. Of particular interest is to estimate quantiles of such distributions, which e.g. in the earth sciences are called return levels. A research field at present is to find suitable models for handling covariates and time dependency. Collaboration has been made with Swedish Radiation Safety Authority, concerning extreme sea levels. 

For further information, contact Jesper Rydén

Advanced statistical modelling and artificial intelligence (AI) with applications across human and animal health, and agriculture science is the focus of Reza Belaghi’s research. He develops AI prediction models for early diagnosis of diseases such as preterm birth in humans, cancer in human (based on miRNAs), cancer in dogs, and digital dermatitis in dairy cattle, as well as for crop yield forecasting and precision farming. 

His methodological expertise includes high-dimensional data theory (p>>n), penalized and shrinkage estimation, zero-inflated models, factor analysis, and censored data (time-to-event analysis), tackling challenges in complex, real-world datasets. Currently at SLU, he integrates domain knowledge with AI/ML to create scalable, interpretable solutions that foster data-driven innovation across disciplines. 

For further information, contact Reza Belaghi