Login for PhD students/staff at UCPH      Login for others
Machine Learning Methods in Non-Life Insurance
Provider: Faculty of Science

Activity no.: 5593-23-07-31 
Enrollment deadline: 01/02/2023
PlaceDepartment of Mathematical Sciences
Universitetsparken 5, 2100 København Ø
Date and time06.02.2023, at: 08:00 - 14.04.2023, at: 16:00
Regular seats50
ECTS credits7.50
Contact personNina Weisse    E-mail address: weisse@math.ku.dk
Enrolment Handling/Course OrganiserJostein Paulsen    E-mail address: jostein@math.ku.dk
Written languageEnglish
Teaching languageEnglish
Semester/BlockBlock 3
Scheme groupB
Exam requirementsTwo mandatory assignments must be approved and valid before the student is allowed attending the exam.
Exam formWritten assignment
Exam details12 hour take-home exam. Collaboration not allowed.
Exam aidsAll aids allowed
Grading scale7 point grading scale. For PhD students: Passed / Not Passed
Internal/external examiner
Censorship formOne external examiner.
Exam re-examinationSame as ordinary exam. If the the two mandatory homework assignments were not approved before the ordinary exam they must be handed in at the latest two weeks before the beginning of the re-exam week. They must be approved before the re-exam.
Course workload
Course workload categoryHours
Exam12.00
Preparation124.00
Lectures28.00
Project work42.00
Sum206.00
Content
Basic theory of penalized linear regression, splines, additive models, neural networks, multivariate adaptive splines, projection pursuit regression, regression trees, random forests, boosting. Also various methods of classification.

Learning outcome
Knowledge:
- Standard penalized methods such as ridge regression and the lasso
- Know splines, additive, additive models, neural networks, MARS
- Regression trees, random forest, boosting
- Classification with classical methods as well as machine learning methods

Skills:
- A general ability to use machine learning methods to solve practical problems

Competences:
- Know how to use R to solve practical problems

Literature
Lecture notes

Teaching and learning methods
4 hours of lectures per week for 7 weeks

Search
Click the search button to search Courses.


Course calendar
See which courses you can attend and when
JanFebMarApr
MayJunJulAug
SepOctNovDec



Publication of new courses
All planned PhD courses at the PhD School are visible in the course catalogue. Courses are published regularly.