Skip to main content Accessibility help

Real-time Bayesian non-parametric prediction of solvency risk

  • Liang Hong (a1) and Ryan Martin (a2)


Insurance regulation often dictates that insurers monitor their solvency risk in real time and take appropriate actions whenever the risk exceeds their tolerance level. Bayesian methods are appealing for prediction problems thanks to their ability to naturally incorporate both sample variability and parameter uncertainty into a predictive distribution. However, handling data arriving in real time requires a flexible non-parametric model, and the Monte Carlo methods necessary to evaluate the predictive distribution in such cases are not recursive and can be too expensive to rerun each time new data arrives. In this paper, we apply a recently developed alternative perspective on Bayesian prediction based on copulas. This approach facilitates recursive Bayesian prediction without computing a posterior, allowing insurers to perform real-time updating of risk measures to assess solvency risk, and providing them with a tool for carrying out dynamic risk management strategies in today’s “big data” era.


Corresponding author

*Correspondence to: Liang Hong, Department of Mathematics, Robert Morris University, Moon Township, PA 15108-2574, USA. Tel: +412 397 4024. E-mail:


Hide All
Brazauskas, V. & Kleefeld, A. (2011). Folded and log-folded-t distributions as models for insurance loss data. Scandinavian Actuarial Journal, 1, 5974.
Brazauskas, V. & Kleefeld, A. (2014). Author’s reply to “Letter to Editor: Regarding Folded and the Paper by Brazauskas and Kleefeld” by Scollnik. Scandinavian Actuarial Journal, 8, 753757.
Brazauskas, Y. & Kleefeld, A. (2016). Modeling severity and measuring tail risk of Norwegian fire claims. North American Actuarial Journal, 20(1), 116.
Bühlmann, & Gisler, (2005). A Course in Credibility Theory and Its Application. Springer, , New York.
Calderín-Ojeda, E. & Kwok, C.F. (2016). Modeling claims data with composite Stoppa models. Scandinavian Actuarial Journal, 9, 817836.
Cooray, K. & Cheng, C.I. (2015). Bayesian estimators of the lognormal-Pareto composite distribution. Scandinavian Actuarial Journal, 6, 500515.
Ferguson, T.S. (1973). Bayesian analysis of some nonparametric problems. Annals of Statistics, 1, 209230.
Frees, E.W., Derrig, R.A. & Meyers, G. (2014). Predictive Modeling Applications in Actuarial Science, Vol. I: Predictive Modeling Techniques. Cambridge University Press, Cambridge.
Fröhlich, A. & Weng, A. (2015). Modeling parameter uncertainty for risk capital calculation. European Actuarial Journal, 5, 79112.
Gerrard, R. & Tsanakas, A. (2011). Failure probability under parameter uncertainty. Risk Analysis, 31, 727744.
Ghosal, S. (2010). The Dirichlet process, related priors and posterior asymptotics. In N.L. Hjort, C. Holmes, P. Müller & S.G. Walker (Eds.), Bayesian Nonparametrics (pp. 35–79). Cambridge University Press, Cambridge.
Hahn, P.R., Martin, R. & Walker, S.G. (2017). On recursive Bayesian predictive distributions. Journal of the American Statistical Association,
Hong, L. & Martin, R. (2017a). A flexible Bayesian nonparametric model for predicting future insurance claims. North American Actuarial Journal, 21(2), 228241.
Hong, L. & Martin, R. (2017b). Dirichlet process mixture models for insurance loss data, Scandinavian Actuarial Journal,
Kalli, M., Griffin, J.E. & Walker, S.G. (2011). Slice sampling mixture models. Statistical Computing, 21, 93105.
Klugman, S.A., Panjer, H.H. & Willmot, G.E. (2012). Loss Models: From Data to Decisions, 4th edition. Wiley, Hoboken, NJ.
Makov, U.E. (2001). Principal applications of Bayesian methods in actuarial science. North American Actuarial Journal, 5(4), 5357.
Martin, R. & Ghosh, J.K. (2008). Stochastic approximation and Newton’s estimate of a mixing distribution. Statistical Science, 23, 365382.
Martin, R. & Tokdar, S.T. (2009). Asymptotic properties of predictive recursion: robustness and rate of convergence. Electronic Journal of Statistics, 3, 14551472.
Martin, R. & Tokdar, S.T. (2011). Semiparametric inference in mixture models with predictive recursion marginal likelihood. Biometrika, 98, 567582.
Martin, R. & Tokdar, S.T. (2012). A nonparametric empirical Bayes framework for large-scale multiple testing. Biostatistics, 13, 427439.
Müller, P. & Quintana, F.A. (2004). Nonparametric Bayesian data analysis. Statistical Science, 19, 95110.
Nadarajah, S. & Bakar, S.A.A. (2015). New folded models for the log-transformed Norwegian fire claim data. Communications in StatisticsTheory and Methods , 44, 44084440.
Own Risk and Solvency Assessment (2017). Available online at the address [accessed on 8-Aug-2017].
Nelson, R.B. (2006). An Introduction to Copulas, 2nd edition. Springer, New York.
Newton, M. (2002). On a nonparametric recursive estimator of the mixing distribution. Sankhyā: The Indian Journal of Statistics , 64, 306322.
Newton, M. & Zhang, Y. (1999). A recursive algorithm for nonparametric analysis with missing data. Biometrika, 86(1), 1526.
Rytgaard, M. (1990). Estimation in the Pareto distribution. ASTIN Bulletin, 20, 201216.
Scollnik, D.P.M. (2001). Actuarial modeling with MCMC and BUGS. North American Actuarial Journal, 5(2), 96124.
Scollnik, D.P.M. & Sun, C. (2012). Modeling with Weibull–Pareto models. North American Actuarial Journal, 16, 260272.
Scollnik, D.P.M. (2014). Letter to editor: regarding folded models and the paper by Brazauskas and Kleefeld (2011). Scandinavian Actuarial Journal, 2014(3), 278281.
Sheather, S.J. (2004). Density estimation. Statistical Science, 19(4), 588597.
Sklar, M. (1959). Fonctions de répartition á n dimensions et leurs marges. Université Paris, 8, 229–231.
Solvency II (2009). Available online at the address [accessed 8-Aug-2017].
Tokdar, S.T., Martin, R. & Ghosh, J.K. (2009). Consistency of a recursive estimate of mixing distributions. Annals of Statistics, 37, 25022522.


Related content

Powered by UNSILO

Real-time Bayesian non-parametric prediction of solvency risk

  • Liang Hong (a1) and Ryan Martin (a2)


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed.