Bayesian Basis Expansion Models

  • Guangyuan GaoEmail author


In this chapter, Bayesian basis expansion models are used to fit various development patterns and accommodate the tail factor. A parametric model is typically characterized by a parametric mean function and an error distribution. The shape of the mean function is restricted by the space of parameters. Non-parametric models such as basis expansion models are able to automatically adjust to fit any shape of data. In Sect. 5.1, the aspects of splines are reviewed, including spline basis functions, smoothing splines, low rank smoothing splines and Bayesian shrinkage splines. In Sect. 5.2, we study two simulated examples. The first simulated example is based on a trigonometric mean function, while the second simulated example is based on the claims payments process. Both examples illustrate the usefulness of natural cubic spline basis in the extrapolation beyond the range of data. Section 5.3 is the application of above methodology to the doctor benefit in WorkSafe Victoria. The basis expansion model used to fit the PPCI triangle induces a tail development.


  1. Armagan, A., Dunson, D. B., & Lee, J. (2013). Generalized double Pareto shrinkage. Statistica Sinica, 23, 119–143.MathSciNetzbMATHGoogle Scholar
  2. Bishop, C. M. (2006). Pattern recognition and machine learning. New York: Springer.zbMATHGoogle Scholar
  3. Crainiceanu, C. M., Ruppert, D., & Wand, M. P. (2005). Bayesian analysis for penalized spline regression using WinBUGS. Journal of Statistical Software, 14, 1–14.CrossRefGoogle Scholar
  4. DiMatteo, I., Genovese, C. R., & Kass, R. E. (2001). Bayesian curve-fitting with free-knot splines. Biometrika, 88, 1055–1071.MathSciNetCrossRefGoogle Scholar
  5. England, P. D., & Verrall, R. J. (2001). A flexible framework for stochastic claims reserving. Proceedings of the Casualty Actuarial Society, 88, 1–38.Google Scholar
  6. Faraway, J. J. (2015). Linear models with R (2nd ed.). Boca Raton: Chapman & Hall.Google Scholar
  7. Gao, G., & Meng, S. (2018). Stochastic claims reserving via a bayesian spline model with random loss ratio effects. ASTIN Bulletin, 48, 55–88.MathSciNetCrossRefGoogle Scholar
  8. Gabrielli, A., Richman, R., & Wüthrich, M. V. (2018). Neural network embedding of the over-dispersed Poisson reserving model. SSRN, ID 3288454.Google Scholar
  9. Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2014). Bayesian data analysis (3rd ed.). Boca Raton: Chapman & Hall.Google Scholar
  10. Hall, P., & Opsomer, J. D. (2005). Theory for penalised spline regression. Biometrika, 92, 105–118.MathSciNetCrossRefGoogle Scholar
  11. Hastie, T. J., Tibshirani, R. J., & Friedman, J. H. (2009). The elements of statistical learning: Data mining, inference, and prediction (2nd ed.). New York: Springer.Google Scholar
  12. Hastie, T. J., & Tibshirani, R. J. (1990). Generalized additive models. New York: Chapman & Hall.zbMATHGoogle Scholar
  13. James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. New York: Springer.CrossRefGoogle Scholar
  14. Komaki, F. (2006). Shrinkage priors for Bayesian prediction. The Annals of Statistics, 34, 808–819.MathSciNetCrossRefGoogle Scholar
  15. Lay, D. C. (2012). Linear algebra and its application (4th ed.). Boston: Addison Wesley.Google Scholar
  16. Park, T., & Casella, G. (2008). The Bayesian lasso. Journal of the American Statistical Association, 103, 681–686.MathSciNetCrossRefGoogle Scholar
  17. Ruppert, D. (2012). Selecting the number of knots for penalized splines. Journal of Computational and Graphical Statistics, 11, 735–757.MathSciNetCrossRefGoogle Scholar
  18. Ruppert, D., Wand, M. P., & Carroll, R. J. (2003). Semiparametric regression. New York: Cambridge University Press.CrossRefGoogle Scholar
  19. Wood, S. N. (2003). Thin plate regression splines. Journal of the Royal Statistical Society B, 65, 95–114.MathSciNetCrossRefGoogle Scholar
  20. Wood, S. (2006). Generalized additive models: An introduction with R. New York: Chapman & Hall.CrossRefGoogle Scholar
  21. Wüthrich, M. V., & Buser, C. (2018). Data analytics for non-life insurance pricing. SSRN, ID 2870308.Google Scholar
  22. Zhang, Y. W., & Dukic, V. (2013). Predicting multivariate insurance loss payments under the Bayesian copula framework. Journal of Risk and Insurance, 80, 891–919.CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  1. 1.School of StatisticsRenmin University of ChinaBeijingChina

Personalised recommendations