NMAK22003U Empirical Bayes and Generalized Linear Mixed Models *CHANGES TO WORKLOAD AND EXAM

Volume 2022/2023
Education

MSc Programme in Actuarial Mathematics
MSc Programme in Mathematics-Economics

Content

Empirical Bayes methods; Conjugate families; Compound models with left truncated data; Linear mixed models (LMM); Generalized linear mixed models (GLMM); Hierarchical generalized linear mixed models (HGLM).

Learning Outcome

The aim of the class is to show how standard and not so standard statistical models can be extended to include random parameters. In insurance this is useful when policyholders can naturally be segmented into groups, so that policies within a group are not independent. For actual solutions of these models numerical integration is often required, but sometimes analytical results are available and we shall pursue some of these. During the course the students will use R programs with some data from insurance. Some  more basic programming will also be required.

Notes written by the lecturer

Regression; Topics in non life insurance
4 hours of lectures and 2 hours of exercises a week for 7 weeks
  • Category
  • Hours
  • Lectures
  • 21
  • Preparation
  • 152
  • Exercises
  • 21
  • Exam
  • 12
  • Total
  • 206
Collective
Credit
7,5 ECTS
Type of assessment
Oral examination, 30 minutes, no preparation time
Exam registration requirements

To be allowed to take the exam two compulsory homeworks must be approved. These can be done in groups.

Aid
Without aids
Marking scale
7-point grading scale
Censorship form
No external censorship
One internal examiner
Re-exam

Oral reexam. Questions not known in advance. If the two mandatory homework assignments were not approved before the ordinary exam they must be resubmitted. They must be approved three weeks before the beginning of the re-exam week.

Criteria for exam assesment

See Learning Outcome.