|Fees||Look up fees|
This subject aims to provide students with basic training on modern data analytics methods, which include linear regression, classification, resampling methods, spline-based methods, generalised additive models and tree-based methods. This subject focuses on applying the above methods to modelling non-life insurance claims frequency and severity.
Intended learning outcomes
On completion of this subject, students should be able to:
- Recognise major types of non-life insurance data and their main characteristics.
- Demonstrate a depth of knowledge in linear regression methods, regression splines and smoothing splines.
- Demonstrate basic understanding in various statistical learning models that include classification methods, re-sampling methods, generalised additive models and tree-based methods.
- Use relevant computer software to apply various statistical learning models for insurance related applications.
- Apply basic linear model selection and regularisation techniques when conducting linear regression analyses and use bagging and boosting techniques to improve decision trees when applying tree-based methods.
- Interpret the results of data analytics conducted on real insurance data
- Compare benefits/drawbacks of competing models and methods, relevant to real problems.
- High level of development: written communication; logical problem solving; statistical reasoning; application of theory to practice; interpretation and analysis; synthesis of data and other information; evaluation of data and other information; use of computer software.
Last updated: 2 December 2019