Workshops and seminar series

Bayesian vs Frequentist Inference: Concepts, Contrasts, and Applications

28 August, Mohammad Ghorbani, SLU.

This introductory session examines the philosophical and practical differences between Bayesian and frequentist approaches to statistical inference. We discuss how these frameworks influence model formulation, parameter interpretation, and uncertainty quantification. Through applied examples, we highlight key decisions in the modeling process and demonstrate how Bayesian methods are used in contemporary research. In addition, the talk will briefly introduce commonly used R packages for Bayesian data analysis and illustrate their role in practical implementation. The seminar is designed to be accessible to a broad audience and serves as a foundation for the subsequent talks in the series.

From snow leopards to house crickets: how a Bayesian approach helps me answer questions in a better way

4 September, Matthew Low, SLU. 

Often I build models in a Bayesian framework that don’t differ in their structure from what I could build more quickly using a likelihood-based R package. So why do I do it? Is it because I like to make my life more complicated? Is it to show off my ‘cool' credentials in being a stats nerd? In this seminar I will show examples from my work in analysing snow leopard prey selection, bird trends across the Great Barrier Reef, and personality traits of insects, and how building my models in a Bayesian framework allows me to ask new questions and answer old ones in a much clearer way. Basically, I use Bayes because it helps me communicate my results.

From ecological data to management decisions: using Bayesian models in carnivore research

18 September, Malin Aronsson, SLU.

A large part of my research focuses on large carnivores, often in situations where ecological uncertainty meets very practical management decisions. In many cases, managers are not primarily asking whether an effect is “significant” or not – they want to know the probability that a management action will lead to a certain outcome. In wildlife management, decisions often need to be made despite substantial uncertainty and incomplete information. Bayesian approaches provide a useful framework for combining ecological data with management goals, and for presenting results in ways that are relevant for practical decision-making. 

In this seminar I will talk about how I use Bayesian models to translate ecological data into directly interpretable probabilities that can be used in both research and wildlife management. I will show examples from work ranging from applied wildlife management to broader ecological questions. Bayesian approaches have helped us develop practical decision-support tools for management authorities, while also improving our understanding of ecological processes. One thing I am particularly interested in is how probabilistic thinking can help bridge the gap between ecological research and applied management of large carnivores.

A Bayesian approach to analyzing the long-term soil fertility experiment

25 September, Rong Lang, SLU.

The treatment effects in the agricultural long-term experiments (LTEs) have been widely analyzed using conventional Frequentist approaches. In contrast, the Bayesian approach based on Bayes’ Theorem incorporates prior information on model parameters and provides greater flexibility on the models and parameters. The aim of this talk is to show how the Bayesian approach can be applied to the LFTs datasets and how results and interpretations differ between the two approaches, using the long-term soil fertility experiment in Sweden as an example. The effects of crop rotation and fertilization on soil carbon (SOC) content in 0-20 cm topsoil were assessed using both the conventional mixed-effects model and a Bayesian approach. The results showed similar trends that rotation with ley slowed SOC loss compared to rotation without ley, and nitrogen fertilization reduced SOC losses.

A Bayesian approach to projecting forest dynamics and related uncertainty: An application to continuous cover forests

2 October, Mari Myllymäki, Luke.

Continuous cover forestry (CCF) is forest management based on ecological principles and this management type is currently re-visited in many countries. CCF woodlands are known for their structural diversity in terms of tree size and species and forest planning in CCF needs to make room for multiple forest development pathways as opposed to only one management target. As forest management diversifies and management types such as CCF become more common, models used for projecting forest development need to have a generic and flexible bottom-up design. They also need to be able to handle uncertainty to a larger extent and more comprehensively than is necessary with single, traditional forest management types.

In this study, a spatial tree model was designed for analyzing a data set involving 18 plots from CCF stands in Southern Finland. The tree model has specific ingrowth, growth and mortality model components, each including a spatially explicit competition effect involving neighboring trees. Approximations were presented that allow inference of the model components operating in annual steps based on time-series measurements from several years. We employed Bayesian methodology and posterior predictive distributions to simulate forest development for short- and long-term projections.

The Bayesian approach allowed us to incorporate uncertainties related to model parameters in the projections, and we analyzed these uncertainties based on three scenarios: (1) known plot and tree level random effects, (2) known plot level random effects but unknown tree level random effects, and (3) unknown random effects. Our simulations revealed that uncertainties related to plot effects can be rather high, particularly when accumulated across many years whilst the length of the simulation step only had a minor effect. As the plot and tree effects are not known when tree models are applied in practice, in such cases, it may be possible to significantly improve model projections for a single plot by taking one-off individual-tree growth measurements from the plot and using them for calibrating the model. Random plot effects as used in our tree model are also a way of describing environmental conditions in CCF stands where other traditional descriptors based on stand height and stand age fail to be suitable any more.

The Workshop was given on the 5th of May 2026.

Scientific papers with systematic reviews have become more common and rigorous over the past couple of decades. This workshop will give an introduction to meta-analysis used in review papers, hands-on exercises using packages in R, we will explore how to visualize the results (using forest plots and funnel plots), and how to interpret the statistics behind meta-analysis. Designed for researchers, PhD students from the Swedish Agricultural University, the workshop will provide both theoretical foundations and practical applications using R software. Prior knowledge of basic statistics (e.g. linear regression) and experience of using R is recommended. Experience of applying linear mixed models is also useful but not a prerequisite.

Target audience

Targeted towards applied researchers and doctoral students. This online workshop will provide both theoretical foundations and practical applications using R software. Prior knowledge of basic statistics (e.g. linear regression) and experience of using R is recommended. Experience of applying linear mixed models is also useful but not a prerequisite.

Workshop leaders

Lars Rönnegård and Adam Flöhr.

Program

• Introduction to meta-analysis and some example review papers
• Exercises using the meta and metafor packages in R
• Interpretation of plots and statistics produced by the packages
• Interpretation of the fitted models within a linear regression and linear mixed model framework
• Questions, Answers and Discussions
• Summary and conclusion

February 13th - 14th, Reza Belaghi: Two-Day Workshop on Unsupervised Machine Learning with Applications in Agricultural Science Using R

This workshop will explore essential techniques in unsupervised learning, focusing on factor analysis, principal component analysis (with application in high-dimensional regression), and clustering methods, with their applications in agriculture, natural, and animal sciences. Designed for researchers, PhD students from the Swedish Agricultural University, the workshop will provide both theoretical foundations and practical applications using R software. 

Participants will gain hands-on experience with real-world datasets, learning how to analyze complex data, reduce dimensionality, and uncover patterns using unsupervised AI. By the end, attendees will be equipped to apply these techniques in their own research fields. 

Prior knowledge of basic statistics is recommended.

Outline

• Introduction to Unsupervised Learning and its applications in agriculture
• Factor Analysis: Theory, assumptions, and interpretation
• Principal Component Analysis (PCA): Dimensionality reduction and visualization
• Hands-on session: Applying PCA to real datasets
• Clustering techniques: k-means, PAM, hierarchical clustering, and DBSCAN
• Hands-on session: Clustering and interpreting results in real-world applications
• Questions, Answers and Discussions.

May 15th - 16th, Reza Belaghi: Two-Day workshop on Study Design and Sampling Techniques in Natural, Animal and Agricultural Sciences

In the fields of natural, animal and agricultural sciences, effective study design and appropriate sampling techniques are essential for producing valid and reliable research results. One of the critical components of study design is determining the optimal sample size, which ensures adequate power to detect meaningful effects while minimizing costs and resources.

This workshop will provide participants with a comprehensive understanding of study design principles, sampling methods, and sample size calculations with real-world applications in their respective fields.

Outline

• Introduction to Study Design
  Overview of the importance of study design in research, highlighting key components such as research questions, hypotheses, and variables.
• Types of Study Designs
  Discussion on various study designs including observational and experimental approaches, with their respective strengths and weaknesses.
• Importance of Sample Size Calculation
  Exploration of the role of sample size in ensuring research validity, addressing consequences of inadequate sample sizes, and introducing concepts of statistical power and error types.
• Sample Size Calculation Methods
  Detailed methods for calculating sample size based on factors like effect size, variability, desired power level, and significance level, along with an introduction to software tools for these calculations.
• Sampling Techniques
  Examination of probability and non-probability sampling methods, emphasizing their relevance and selection criteria for different research contexts.
• Applications in Natural, Animal, and Agricultural Sciences
  Presentation of case studies and examples that illustrate the application of study design, sampling techniques, and sample size calculations in real-world scenarios.
• Practical Workshop Activities
  Group activities focusing on designing studies and calculating sample sizes for specific research questions, followed by presentations and feedback sessions.
• Conclusion and Q&A
  Summary of key concepts and an open forum for questions and discussions.

April 29, Annica de Groote and Peter Lundquist: Design of Questionnaire Surveys.

A survey can be looked upon as a process or series of survey operations. The design of each operation has the potential to affect the total quality of the survey results. Different methods of the way individuals are selected to the study have different strengths and weaknesses, and the choice can have great significance for the quality and credibility of the survey results. Another design choice of interest here is the strategy for dealing with nonresponse - both preventively and when nonresponse has arisen. Nonresponse is large in many surveys and can seriously bias the results.

May 17, Claudia von Brömssen: Geographically weighted regression model – identifying spatially differentiated relationships and trends.

Geographically Weighted Regression (GWR) is a spatial statistical technique to model spatially varying relationships between variables.

 
May 23-24, Reza Belaghi: Advanced Regression Analysis in Natural Sciences with R Software.

Through hands-on exercises and real-world examples, this workshop aims to equip attendees with a comprehensive skill set in advanced regression analysis, providing practical insights into modeling techniques for count data, survival analysis, and addressing issues related to excess zeros in the data.

June 12-13, Lars Rönnegård: Generalized Linear Mixed Models with extensions using R.

Generalized linear mixed models are GLMs with random effects. This is a class of models allowing non-normal outcomes and dependencies between observations with applications in analysis of repeated observations, spatial data and genetics.

Introduction to Machine Learning in Natural Science: Modeling and Applications

November 29-30, Reza Belaghi.

In this workshop, we will explore the application of state-of-the-art machine learning models in the field of natural science, using real-world examples and various data sets. Our goal is to equip participants with the necessary knowledge and skills to apply machine learning in their research and scientific papers, and applications (whenever is needed).

Analysing network data that are partially observed

Johan Koskinen is a Lecturer in Statistics at Stockholms universitet. His main research interests centre on statistical modelling and Bayesian inference for networks. 

Network data may be represented as binary graphs, either directed or undirected, and have a long history of being used to model and describe interaction between people and other entities, with formal approaches dating back to at least the start of the twentieth century.  For a graph, the potential tie between a pair of nodes is represented by a binary indicator variable that we may call a tie-variable. These tie-variables are indexed by the labels of the nodes and can be organised in a so-called node by node adjacency matrix. Since the entries of the adjacency matrix are cross-classified by both the row node, and the column node, the tie-variables are highly interdependent. Exponential (family) random graph models (ERGMs) constitute a class of log-linear models with natural parameters that have as statistics a subset of graph statistics derived out of principled dependence assumptions. Due to these dependencies, the ERGM for a network does not marginalise and subgraphs of the network do not follow models of the same form. Here we discuss inference approaches for the parameters of the ERGM when some tie-variables are missing. The treatment of missing data in ERGM also applies to cases where data are missing by design, for example when the network data have been obtained through a link-tracing designs, such as snowball sampling. We describe a Bayesian approach for estimation and provide examples of applications to networks of young men who have sex with men, rebels in the Democratic Republic of the Congo, as well as the use of the Bayesian estimation scheme for imputing initial conditions in the analysis of network panel data. The latter case is illustrated with an application to social support networks in bushfire-affected communities in Australia. The use of the proposed approach is contingent of a number of fairly heroic assumptions, some of which will be brought up for discussion.

 Seminar series: Survey methodology

September 16, Anton Grafström, Department of Forest Resource Management: Spatially balanced sampling for environmental monitoring

September 23, Eros Quesada, Department of Aquatic Resources: Anomaly detection techniques applied to fishery data for the identification of possible misreporting

September 30, Petter Kjellander, Department of Ecology – Distance sampling in wildlife management

October 7, Peter Lundquist, Department of Energy and Technology: Jackknife variance estimation for a complex survey of land use

Seminar series: Environmental Statistics

August 27, Claudia von Brömssen, SLU: Statistical methods for evaluation of temporal trends in environmental data. 

September 10, James Weldon, SLU:  Change, stability and atmospheric pollutant effects in European forest vegetation (Defense of doctoral thesis)

September 17, Xin Zhao, SLU: Design-based sampling methods for environmental monitoring (Defense of doctoral thesis)

September 24, Arne Pommerening, SLU: Individual-based tree modelling for remote sensing data.

October 1, Jesper Rydén, SLU: Modelling extreme values: problems and concepts

October 15, Martin Sköld, Naturhistoriska riksmuseet, Disentangling effort and density in non-invasive genetic sampling by volunteers, the case of the Swedish Brown Bear monitoring programme

October 22, Annica de Groote, SLU: An introduction to sampling for natural resources

October 29, Anders Grimvall, Havsmiljöinstitutet and Linköpings university: How far can the evaluation of monitoring data be automated?

Seminar serie: The digital SLU

August 28: Dorota Anglart, SLU & DeLaval: Generalized additive model for dairy cow somatic cell count predictions using sensor data

September 4: Aakash Chawade, Dept of Plant Breeding, SLU: Challenges and opportunities for analysis of omics data

September 11: Mats Söderström & Kristin Piikki. Dept. of soil and environment: Spatial data for mapping of crop and soil characteristics: Digital soil mapping, modelling crop status from remote sensing data, data fusion, multi-scale modelling, sampling strategies, validation

September 18: Keni Ren, Umeå University: Zoom in on the precision livestock farming

September 25: Måns Thulin, consultant in statistics and AI: An introduction to statistical learning

October 2: Johanna Bergman, AI innovations of Sweden: AI Innovation of Sweden and SLU

October 9: Moudud Alam, Dalarna University: Monitoring reindeer activities in their natural environment

October 16, Johan Holmgren, Dept. for forest resource management, SLU: Forest remote sensing on the individual tree level 

October 23: Bo Stenberg & Johanna Wetterlind, Dept. of soil and environment: Spectral data from proximal sensors for analysis of soil properties: Instruments, data collection, data preparation, modelling, validation

Workshop in Machine Learning

The term machine learning hides a variety of different statistical methods. The basic aim of these methods is to recognize or disclose structures / patterns in data. What is often sold a bit like highly complex mathematics usually has simple ideas as a basis.

Some of the theoretical ideas behind machine learning will be presented. However, the focus is on the implementation of these methods in R and the interpretation of the results. In the first part of the workshop, we will deal with continuous response variables, whereas in the second part we will work with category data.

We would like to invite Ph.D. students and researchers to a two-part workshop in R machine learning given by Sven Adler. Part one on 4-5 november (13:00-16:00 and 9:00-12:00), part two on 14-15 december (13:00-16:00 and 9:00-12:00) in Umeå.

Workshop in Generalized Additive Models

Claudia von Brömssen: Introduction to General Additive Models

Claudia von Brömssen: Generalised and mixed models in the GAM context

Michal Zmihorski: GAM in ornothological studies

Henrik Thurfjell: GAM in modelling bear populations

Sven Adler: GAM in species habitat modelling

Stefan Widgren: GAM for modelling the prevalence of infectious diseases

Mikael Franko: GAM for modelling excess mortality of infectious diseases

Valerio Bartolino: GAM in fish ecology

Willem Dekker: GAM for modelling the annual recruitment data of young eels

Jens Fölster: GAM for trend analysis in water quality data

Claudia von Brömssen: GAM for modelling time-varying relationships

Workshop in Mixed models

Ulf Olsson: Mixed Models (first day)

Lin Shi, Food Science: The application of mixed model in exploring time dependent postprandial metabolic changes - a randomized, cross-over study (second day)

Andrew Allen, Wildlife, Fish and Environmental Studies: Understanding intraspecific variation in movement patterns of moose: A multi-scale approach (second day)

Wiebke Neumann, Wildlife, Fish and Environmental Studies: Using mixed models to analyze autocorrelated data in nested design. – Examples from the analyses of moose GPS positions across species' latitudinal range (second day)

Johan Pihel, Landscape Architecture, Planning and Management: Mixed effect models in Eye tracking and visual assessment studies of forest landscapes  (second day)

Johannes Forkman, Crop Production Ecology: Randomized block trials with spatial correlation (second day)

Workshop in generalised linear models

Ulf Olsson: Generaliserade linjära modeller

Ulf Olsson/Jonas Oliva Palau: Pseudo-binomial data

Johannes Forkman: Overdispersion

Workshops in mixed models

Ulf Olsson: Mixed Models (first day)

Jan-Eric Englund: Does it matter if you use Mixed Models? (second day)

Johannes Forkman: Randomised block trials with spatial correlation (second day)