Workshops

Last changed: 19 February 2018
graftree.jpg

Individual/Agent-Based-Modelling

The Hands-On Workshop is given  20/21 August 2018 at SLU Umeå, Sweden.

Agent/Individual-based modelling (ABM, IBM) techniques have emerged as crucial tools for analysing and understanding self-organisation processes in various ecosystems as well as in social sciences. ABM/IBM is a methodology for relating the micro-level behaviour to the macro- or system-level behaviour and shares certain similarities with game theory. In combination with new methods of spatial statistics ABM/IBM have become an intriguing method for studying spatio-temporal patterns among other things.

An important part of individual-based models are interaction processes between individuals or agents, e.g. trees, humans or animals. For describing and modelling interaction, IBM modellers mostly use kernel functions. This concept has its origin in various fields of natural sciences such as in the ecological field theory and in shot-noise fields of physics. These interaction functions describe how biological processes such as growth, survival and reproduction of an individual depend on its own size and the size of and distance to other individuals. In the workshop, we will also consider and discuss birth- and death processes including the way how they are related with interaction.

A two-days practical workshop will be held at Umeå where beginners and more advanced modellers in the realm of agent/individual-based modelling mix and meet. The objective of the workshop is to share knowledge, experience and tips and tricks in this important field of modelling and to advance our understanding of how to apply this concept to various fields of ecology. Following 1-3 keynote presentations the participants present their work or state their interest. During these presentations burning issues and problems are identified, discussed and possible solutions exchanged, e.g. the statistical estimation of model parameters, programming issues and model functions. In this workshop, groups with similar interest work on possible solutions to their problems and the results are presented in a final plenary session. Participants are encouraged to bring draft manuscripts or sketches of initial ideas for papers. Ideally – as a long-term outcome of the workshop – several papers are published on this topic by the participants.

The preliminary programme includes Dr. Alejandra Vovides, University of Glasgow, UK and Dr. Pavel Grabarnik, Russian Academy of Sciences, Pushchino.

 

This Workshop in Individual/Agent-Based-Modelling is an initiative of IUFRO group 4.01.06 – “Quantitative Ecology” in collaboration with the Graduate School in Applied Statistics & Scientific Computing at the Swedish University of Agricultural Sciences.

Fee: We feel committed to run this workshop at low costs. Therefore we raise only €50 per person to cover the costs of food including a workshop dinner.

Register: Arne Pommerening, SLU Umeå, Sweden. 

 

 

Statistics@SLU gives  workshops on statistical modelling topics.

Do you have any suggestions for forthcoming workshops? 

The workshops are part of the Research School in Applied Statistics and Scientific Computing. If you have suggestions for forthcoming topics contact a statistician at your campus or the Centre (statistics@slu.se).

2015 - Mixed models

Workshop in Mixed models

Presentations:

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)


Exercises:

Forest data:

 

Dogs data:

 

Ladybirds data:

 

Bayisa data:

 

2014 - Generalised linear models

Workshop in generalised linear models  

Föreläsningar

Ulf Olsson: Generaliserade linjära modeller

Ulf Olsson/Jonas Oliva Palau: Pseudo-binomial data

Johannes Forkman: Overdispersion

Johannes Forkman: Overdispersion - code and references

Data sets for computer labs:

Exercise 1: Lichen data, logistic regression

Description

Data as Excel-file

Data as SAS-program code

SAS solutions

 

Exercise 2 and 3: Butterfly data, Poisson regression

Description

Data as Excel-file

Data as SAS-program code

SAS solutions - exercise 2

 SAS solutions - both exercises

 

Exercise 4: Sawfly data, logistic mixed model

Description

Data as Excel-file

Data as SAS-program code

 SAS solutions

 

Exercise 5: Cod data, zero-inflated poisson regression

Description

Data as .txt-file

Data as SAS-program code

SAS solutions

R solutions

 

2013 - Mixed models

Workshop in Mixed Models

Presentations:

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 korrelation (second day)


Exercises:

Forest data:

Dogs data:

Ladybirds data:

 

Bayisa data:


Contact

Arne Pommerening, Professor
Department of Forest Resource Management/Division of Forest Resource Analyses, SLU
arne.pommerening@slu.se, 090-786 82 47, 070-226 97 46