This event has now ended. For further information regarding previous or upcoming events, please email firstname.lastname@example.org
Actuarial Data Science Training: General Insurance (Pricing and Profitability)
3 day Training Event London
We are pleased to present a 3 day hands on general insurance training event in London.
The aim of this workshop is to:
- Present basic and more advanced actuarial and statistical techniques used in non-life pricing, competition analysis and profitability analysis.
- Explore practical problems faced by pricing actuaries and product managers by working through examples and case studies
- Introduce machine learning techniques used in non-life pricing in order to open new perspectives for product development (for example competition analysis and profitability analysis).
Participation is flexible: you can sign up for Day 1, Day 2, Day 3 or all days according to your needs and availability. Please see below a brief outline of each training day with a link to the relevant event and additional course information.
The event is hosted by Actuartech and Reacfin.
The event will count towards CPD.
(This session could be useful for life actuaries looking for an introduction to GLM)
08:30 - Registration
9h-10h30 − Introduction to risk classification
− From linear to generalized linear models
− Poisson regression for claim counts
10h45-12h30 − Case study: Developing a new technical
tariff for frequency
13h30-15h30 − Gamma regression for attritional claims
− Extreme value theory for large claims
− Case study: Developing a new technical
tariff for cost
15h45-17h30 − Case study: Final technical tariff
− Other practical difficulties with GLM
9h-10h30 − Modelling continuous explanatory
variables with Generalized additive
models: methodology and examples
− Penalized regression techniques (Lasso,
Ridge, interaction detection,…):
methodology and examples
10h45-12h30 − Introduction to supervised machine
learning algorithms, regression trees &
− Example: Fitting a regression tree and
random forest on frequency
13h30-15h30 − Case Study: Regression tree and random
forest model adjustment for cost
− Gradient Boosting Model (GBM)
− Example: Fitting GBM on frequency
− Case Study: GBM adjustment for cost
15h45-17h30 − Artificial Neural Networks (ANN)
− Example: Fitting ANN on frequency
− Case Study: ANN adjustment for cost
9h-10h30 − Data Management: Selection, PreAnalysis, Feature Engineering and
− Case Study: Data analysis and features
selection with random forest
10h45-12h30 − Case Study: Continuous Variables
categorization using regression trees or
− Case Study: Application of GBM method
to highlight interactions
13h30-15h30 − Profitability and Competition analysis:
profitability and positioning assessment,
reverse engineering of competitors
15h45-17h30 − Introduction to unsupervised machine
learning algorithms (k-means and HAC):
methodology and examples
− Case study: Profitability
Attendees are encouraged to bring a laptop computer with R installed as well as some useful packages (all the information will be provided after subscription). A basic knowledge of the R software will be useful, but not required.
The training will be conducted in English.