105.702 AKFVM AKSTA AKINF Machine Learning Methods and Data Analytics in Risk and Insurance
This course is in all assigned curricula part of the STEOP.
This course is in at least 1 assigned curriculum part of the STEOP.

2018S, VU, 2.0h, 3.0EC, to be held in blocked form
TUWEL

Properties

  • Semester hours: 2.0
  • Credits: 3.0
  • Type: VU Lecture and Exercise

Aim of course

Machine learning and data analytics is an emerging field that is beginning to have a strong influence on the field of actuarial science practice. The onset of big data applications in insurance has driven the profession to explore new ways to understand data and modelling. Unlike in Google and Facebook type technology applications where huge data bases of labelled data are available, in the insurance context we are often considering unsupervised learning methods. This course will address core methodology to tackle unsupervised problems of relevance to insurance applications.

Subject of course

  • Introduction to unsupervised ML with context of insurance
  • Brief statements of axiomatisation of clustering and impossibility theorem results for clustering
  • Preparing data for clustering
  • K-means clustering, K-centroids
    (with examples in cyber risk and insurance)
  • Information theoretic interpretations and Bregmann bias
  • Hard vs. soft assignment methods and probabilistic clustering
  • Expectation-maximization methods for clustering.
  • Frequentist and Bayesian-EM variants
  • Applications of EM algorithm and variants.
  • (claims reserving examples)
  • Feature maps and kernel maps.
  • Non-linear clustering via kernel k-means
    (telematics data)
  • Families of kernels
  • Kernel target alignments and hyper-parameter tuning
  • Feature extraction methods
  • Probabilistic PCA and robust PPCA factor models
  • Mortality modelling examples
  • Un-supervised multi-kernel learning
  • Classification trees and random forests
    (home insurance - aggregators)
  • Ada boost and bagging

Additional information

The speakers will use the statistical software R with the editor RStudio during the lectures. Participants can (but do not have to) bring along their own laptop with the most recent version of R installed and either a good R programmer editor or R IDE (e.g., the open source edition of RStudio).

Lecturers

  • Shevchenko, Pavel
  • Peters, Gareth

Institute

Course dates

DayTimeDateLocationDescription
08:30 - 18:0009.07.2018 - 13.07.2018FH Hörsaal 1 - MWB course on each of the 5 days
Mon18:00 - 20:0009.07.2018FH Hörsaal 1 - MWB Welcome Reception (invitation of all participants including students of TU Wien)
AKFVM AKSTA AKINF Machine Learning Methods and Data Analytics in Risk and Insurance - Single appointments
DayDateTimeLocationDescription
Mon09.07.201808:30 - 18:00FH Hörsaal 1 - MWB course on each of the 5 days
Mon09.07.201818:00 - 20:00FH Hörsaal 1 - MWB Welcome Reception (invitation of all participants including students of TU Wien)
Tue10.07.201808:30 - 18:00FH Hörsaal 1 - MWB course on each of the 5 days
Wed11.07.201808:30 - 18:00FH Hörsaal 1 - MWB course on each of the 5 days
Thu12.07.201808:30 - 18:00FH Hörsaal 1 - MWB course on each of the 5 days
Fri13.07.201808:30 - 18:00FH Hörsaal 1 - MWB course on each of the 5 days
Course is held blocked

Examination modalities

Attendance of all parts of the course is required and will be checked.
Additional requirements (e.g., excercise classes, homework, oral exam) will be announced as soon as possible.

Course registration

Begin End Deregistration end
24.04.2018 00:00 02.07.2018 23:59 09.07.2018 12:00

Registration modalities

The course is announced as "VU" (Vorlesung mit integrierter Übung) - this means it is a lecture with a practical part (exercises).  Parcipation is necessary and there will be an additional homework - every registered student will be graded.

Advanced students of mathematics (Financial and Actuarial Mathematics, Statistics and Mathematics in Economics, Technical Mathematics) will be accepted preferable. But the course might also be interesting for students of informatics (with interest in maths&statistics).

Curricula

Literature

In case the lecturers provide electronic material (e.g., slides, R codes) this will be distributed to the students.  This applies also to material from additional speakers, which are not mentioned at the course.

Miscellaneous

Language

English