Geographic ratemaking with spatial embeddings
From MaRDI portal
Publication:5067876
Abstract: Spatial data is a rich source of information for actuarial applications: knowledge of a risk's location could improve an insurance company's ratemaking, reserving or risk management processes. Insurance companies with high exposures in a territory typically have a competitive advantage since they may use historical losses in a region to model spatial risk non-parametrically. Relying on geographic losses is problematic for areas where past loss data is unavailable. This paper presents a method based on data (instead of smoothing historical insurance claim losses) to construct a geographic ratemaking model. In particular, we construct spatial features within a complex representation model, then use the features as inputs to a simpler predictive model (like a generalized linear model). Our approach generates predictions with smaller bias and smaller variance than other spatial interpolation models such as bivariate splines in most situations. This method also enables us to generate rates in territories with no historical experience.
Recommendations
- Neural network embedding of the over-dispersed Poisson reserving model
- Spatial modelling of risk premiums for water damage insurance
- Geographic Premium Rating by Whittaker Spatial Smoothing
- Territorial risk classification using spatially dependent frequency-severity models
- Spatial Tweedie exponential dispersion models: an application to insurance rate-making
Cites work
- A data driven binning strategy for the construction of insurance tariff classes
- Bayesian Premium Rating with Latent Structure
- Deep learning
- Geographic Premium Rating by Whittaker Spatial Smoothing
- Non-life rate-making with Bayesian GAMs
- Spatial modelling of claim frequency and claim size in non-life insurance
- Territorial risk classification using spatially dependent frequency-severity models
Cited in
(5)- Actuarial applications of word embedding models
- Non-Life Insurance Risk Classification Using Categorical Embedding
- The use of autoencoders for training neural networks with mixed categorical and numerical features
- A Nested GLM Framework with Neural Network Encoding and Spatially Constrained Clustering in Non-Life Insurance Ratemaking
- Matrix-based factor analysis on the prediction of insurance claims probability
This page was built for publication: Geographic ratemaking with spatial embeddings
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5067876)
