How analytics is improving homeowners pricing
An overview of pricing and analytics for home insurance. Differences to and similarities with other lines of business.
Duration: 60 minutes
The COVID-19 pandemic has influenced the interaction between humans and homes: the living space has becomes the centre of people’s life and work, even more so when under financial strain.
The techniques applied to pricing homeowners insurance are inherently very similar to those for one applied to other LoBs (including Motor). The road to making them more sophisticated starts from with risk models, built at peril level, that allow the calculation of the premium that should be charged to every individual policy to make neither a profit nor a loss. These should leverage on all available external data, in combination with Munich Re’s data and the rating parameters that can be engineered via using advanced analytics techniques. Lifetime-based pricing combines risk and behavioural modelling and helps underwriting and pricing decisions.
- General introduction to homeowners insurance, private clients
- How are products built and governance organised?
- Pricing sophistication: how to win in a competitive market
- Generalised Linear Models, a standard practice
- Machine Learning: going beyond the traditional approach
- RE_Chart and RE_Map: feature engineering for homeowners insurance