STK4100 – Introduction to Generalized Linear Models
Schedule, syllabus and examination date
Teaching and exams spring 2022
In light of the most recent infection control regulations, we will at the start of the spring semester 2022 increase our online teaching, while we at the same time try to maintain in-person teaching where this is possible. We hope to go back to more in-person teaching later on in the semester. You will be informed about any changes in teaching or examinations on the semester page, in Canvas or through your regular channels.
Read more about postponed exams for the autumn semester 2021.
The course gives an overview over important models and techniques for regression analysis outside standard linear regression. In particular, the students will learn how to extend the linear model for response variables with binary/binomial, Poisson and gamma distributions. Furthermore, the students will learn about regression methods for dependent response variables.
After completing the course you will:
- be familiar with the exponential family of distributions and know that the normal, the binomial, the Poisson, and the gamma distributions belong to this family
- know the class of generalized linear models (GLM) as regression models with responses from the exponential family of distributions
- be trained in analyzing data from important special cases of GLMs, in particular, logistic regression and Poisson regression
- know the concepts of link functions for modeling the correspondence between the expected value of the responses and covariates and of variance functions for specifying the correspondence between the expected values and variances of the responses
- be familiar with extensions of the GLM framework using quasi-likelihood based on specified link and variance functions
- know extensions of GLMs that enable modeling and analysis of dependent responses, in particular variance component models and mixed models with both fixed and random components.
Admission to the course
Students admitted at UiO must apply for courses in Studentweb. Students enrolled in other Master's Degree Programmes can, on application, be admitted to the course if this is cleared by their own study programme.
Nordic citizens and applicants residing in the Nordic countries may apply to take this course as a single course student.
If you are not already enrolled as a student at UiO, please see our information about admission requirements and procedures for international applicants.
Recommended previous knowledge
- STK1100 – Probability and Statistical Modelling
- STK1110 – Statistical Methods and Data Analysis
- MAT1100 – Calculus
- MAT1110 – Calculus and Linear Algebra
- MAT1120 – Linear Algebra
- 10 credits overlap with STK2000 – Some central models and methods in statistic (discontinued).
- 10 credits overlap with STK3100 – Introduction to Generalized Linear Models.
- 10 credits overlap with ST-IN216.
- 5 credits overlap with STK4900 – Statistical Methods and Applications.
- 5 credits overlap with STK9900 – Statistical Methods and Applications.
5 hours of lectures/exercises per week throughout the semester.
The course may be taught in Norwegian if the lecturer and all students at the first lecture agree to it.
Final written exam which counts 100 % towards the final grade.
This course has 2 mandatory assignments that must be approved before you can sit the final exam.
It will also be counted as one of the three attempts to sit the exam for this course, if you sit the exam for one of the following courses: STK3100 – Introduction to Generalized Linear Models
Examination support material
Approved calculators are allowed. Information about approved calculators in Norwegian.
Language of examination
Courses taught in English will only offer the exam paper in English. You may write your examination paper in Norwegian, Swedish, Danish or English.
Grades are awarded on a scale from A to F, where A is the best grade and F is a fail. Read more about the grading system.
Resit an examination
This course offers both postponed and resit of examination. Read more: