Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/gavinsimpson/au-viborg-gam-course


https://github.com/gavinsimpson/au-viborg-gam-course

Last synced: 23 days ago
JSON representation

Awesome Lists containing this project

README 

        
# Generalized additive modelling with R



### Aarhus University PhD Course



* 2024 running: December 10th – 12th



* ECTS credits: 1.5 ECTS



* Language: English



* Fee: 350 DKK



## Name of course leader



Gavin Simpson, Assistant Professor, Department of Animal and Veterinary Sciences, Aarhus University [email protected]



### Registration



To register for the course, please contact Julie Jensen on [email protected].



## Objectives of the course



The course will provide an applied introduction to generalized additive modelling in R for biologists. Most of the statistical methods you are likely to have encountered will have specified fixed functional forms for the relationships between covariates and the response, either implicitly or explicitly. These might be linear effects or involve polynomials, such as x + x2 + x3. Generalized additive models (GAMs) are different; they build upon the generalized linear model by allowing the shapes of the relationships between response and covariates to be learned from the data using splines. Modern GAMs are a general data analysis framework, encompassing many models as special cases, including GLMs and GLMMs, and the variety of splines available to users allows GAMs to be used in surprisingly large situations. In this course we’ll show you how to leverage the power and flexibility of splines to go beyond parametric modelling techniques like GLMs.



## Learning outcomes and competences



After completing the course, participants will



* understand how GAMs work from a practical viewpoint to learn relationships between covariates and response from the data

* be able to fit GAMs in R using the mgcv package

* know the differences between the types of splines and when to use them in your models

* know how to visualise fitted GAMs and to check the assumptions of the model

* know how to test specific hypotheses and estimate quantities of interest using fitted models,

* be able to use the R statistical software and in particular the *mgcv*, *gratia*, and *marginaleffects* packages to fit and analyse generalized additive models.



## Compulsory programme



Active participation in the course including attendance at lectures and completion of computer-based classes and exercises. Completion of short, computer-based assessments testing their understanding of a topic and the practical skills taught. For credit, students must complete a data analysis exercise to be submitted one week after the end of the course (19th December).



## Course content



The course is based on a series of lectures and computer-based practical classes led by an international expert in generalized additive modelling and who is the author of gratia, an R package for working with GAMs fitted using the mgcv package. 



The course covers the following topics:



* A recap of generalized linear models for data that are not Gaussian

* Fitting GAMs using mgcv

* Working with penalized splines to estimate flexible effects of covariates

* Model diagnostics and assessment

* Estimating marginal effects and adjusted predictions with GAMs

* Hypothesis testing using GAMs

* Displaying model estimates and reporting results



## Prerequisites



This course is suitable for Phd students (including senior thesis-based masters students) and researchers working with biological data who want to fit models that allow for nonlinear relationships (effects) of covariates on responses.  The course will be of particular interest to PhD candidates and researchers in inter alia biology, animal science, ecology, agriculture, and environmental science. Some prior knowledge of R is required, and some prior knowledge of generalized linear modelling in R would be an advantage.