NOVA-413 Mixed-Effect Models for Forest Applications with Examples in R

Credits (ECTS):3

Course responsible:Hans Ole Ørka

Campus / Online:Taught campus Ås

Teaching language:Engelsk

Limits of class size:25

Course frequency:Not taught in 2024.

Teaching and exam period:Teaching spring. Not in 2024.

About this course

Mixed-effect modelling is important for modelling based on data with various kinds of grouped structures. Examples include data from clustered sample plots in forest inventories, longitudinal data where observations have been made on the same objects on repeated occasions, and hierarchical data structures where, e.g., trees on plots are the study objects. Similar data structures are also common in other disciplines than forest inventory. The effect of groups can be modelled either fixed or random. The core of expected learning outcomes is which effect to choose in the modelling approach. This course provides an introduction to the mixed-effect modelling theory starting with a theory on modelling with categorical variables through generalized linear mixed-effect modelling when non-gaussian assumptions are employed. The course gives examples of practical usage and intuition on to how the modelling methods are applied.

Upon successful completion of the course, participants will be able to understand general concepts of regression analysis with grouped data. The course is intended for students and researchers in ecology, natural resources, forestry, agriculture and environmental sciences.

The course will include the pre-course self-study through literature reading, and the post-course home exam.

  • Pre-course assignment 10 hours
  • Working in the classes 35 hours
  • Home exam after the course 30 hours

The expected number of participants in the course is 25 students.

The language is English.

Learning outcome

Knowledge and understanding

  1. outline general concepts of modelling with grouped data sets;
  2. explain a difference between random and fixed effects;
  3. perform regression analysis applying linear mixed-effect models, nonlinear mixed effect models, and generalized linear mixed-effect models.

Skills and abilities

4. regression analysis with grouped data.

  • The course will consist of lectures mixed with exercises. At the end of the course, students will be given the home exam tasks, which should be seen as a repetition of the previously given material.

    • Working with categorical data
    • Linear mixed-effect models
    • Mixed or random effect: which to apply
    • Non-linear mixed-effect models
    • Generalized linear mixed-effect models

    The course will based on R statistical software.

  • Before the course begins, students are expected to spend time on self-study amounting to at least 10 hours through reading Chapters 4 and 5 in "Biometry for forestry and environmental data: With examples in R" by Lauri Mehtätalo,L. Juha Lappi.

    It is recommended that participants have a basic knowledge in regression analysis and statistics, e.g. acquired through participation in a basic course at the Master level.

  • Home exam. The grading for the course is "Pass" or "Fail" and will be based on the evaluation of the home exam.
  • This course is a joint Nordic NOVA PhD course organised by Hans Ole Ørka, NMBU. Please see the course information webpage on NOVA's website for more information on the course.

    The computer exercises will be done in the lecture room using private laptop computers.

    9 - 13 May, 2022. 08.30 - 16.00 each day.

  • Date: 9 - 13 May, 2022. Time: 08.30 - 16.00 each day.
  • Students on PhD level have number one prioritiy. Master's students may inquire vacancy for PhD courses from the contact listed on the course's NOVA webpage.
  • Passed / Not Passed
  • The course is primarily intended for PhD students, but post-doctoral researchers are also welcome to attend.