Skip to main content Accessibility help
×
Hostname: page-component-586b7cd67f-r5fsc Total loading time: 0 Render date: 2024-11-28T06:15:15.301Z Has data issue: false hasContentIssue false

9 - Finite Mixture Model and Workers’ Compensation Large-Loss Regression Mixture Model and Workers’ Compensation Large-Loss Regression Analysis

Published online by Cambridge University Press:  05 August 2016

Luyang Fu
Affiliation:
University of Illinois at Urbana-Champaign
Edward W. Frees
Affiliation:
University of Wisconsin, Madison
Glenn Meyers
Affiliation:
ISO Innovative Analytics, New Jersey
Richard A. Derrig
Affiliation:
Temple University, Philadelphia
Get access

Summary

Chapter Preview. Actuaries have been studying loss distributions since the emergence of the profession. Numerous studies have found that the widely used distributions, such as lognormal, Pareto, and gamma, do not fit insurance data well. Mixture distributions have gained popularity in recent years because of their flexibility in representing insurance losses from various sizes of claims, especially on the right tail. To incorporate the mixture distributions into the framework of popular generalized linear models (GLMs), the authors propose to use finite mixture models (FMMs) to analyze insurance loss data. The regression approach enhances the traditional whole-book distribution analysis by capturing the impact of individual explanatory variables. FMM improves the standard GLM by addressing distribution-related problems, such as heteroskedasticity, over- and underdispersion, unobserved heterogeneity, and fat tails. A case study with applications on claims triage and on high-deductible pricing using workers’ compensation data illustrates those benefits.

Introduction

Conventional Large Loss Distribution Analysis

Large loss distributions have been extensively studied because of their importance in actuarial applications such as increased limit factor and excess loss pricing (Miccolis, 1977), reinsurance retention and layer analysis (Clark, 1996), high deductible pricing (Teng, 1994), and enterprise risk management (Wang, 2002). Klugman et al. (1998) discussed the frequency, severity, and aggregate loss distributions in detail in their book, which has been on the Casualty Actuarial Society syllabus of exam Construction and Evaluation of Actuarial Models for many years. Keatinge (1999) demonstrated that popular single distributions, including those in Klugman et al. (1998), are not adequate to represent the insurance loss well and suggested using mixture exponential distributions to improve the goodness of fit. Beirlant et al. (2001) proposed a flexible generalized Burr-gamma distribution to address the heavy tail of loss and validated the effectiveness of this parametric distribution by comparing its implied excess-of-loss reinsurance premium with other nonparametric and semi-parametric distributions. Matthys et al. (2004) presented an extreme quantile estimator to deal with extreme insurance losses. Fleming (2008) showed that the sample average of any small data from a skewed population is most likely below the true mean and warned the danger of insurance pricing decisions without considering extreme events. Henry and Hsieh (2009) stressed the importance of understanding the heavy tail behavior of a loss distribution and developed a tail index estimator assuming that the insurance loss possess Pareto-type tails.

Type
Chapter
Information
Publisher: Cambridge University Press
Print publication year: 2016

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Anderson, D., S., Feldblum, C., Modlin, D., Schirmacher, E., Schirmacher, and N., Thandi. A practitioner's guide to generalized linear models. Casualty Actuarial Society Discussion Paper Program, 2004.
Beirlant, J., G., Matthys, and G., Dierckx. Heavy tailed distributions and rating. ASTIN Bulletin, 31: 37–58, 2001.Google Scholar
Boucher, J., and D., Davidov. On the importance of dispersion modeling for claims reserving: An application with the Tweedie distribution. Variance, 5(2): 158–172, 2011. Clark, D. Basics of reinsurance pricing. CAS Exam Study Note, Casualty Actuarial Society, 1996.Google Scholar
Clark, D. A note on the upper-truncated Pareto distribution. ERM Symposium, 2013.
Dean, C. G. Generalized linear models. In E. W., Frees, G., Meyers, and R. A., Derrig (eds.), Predictive Modeling Applications in Actuarial Science: Volume 1, Predictive Modeling Techniques, pp. 107–137. Cambridge University Press, Cambridge, 2014.
Dempster, A. P., N. M., Laird, and D. B., Rubin. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Series B, 39(1): 1–38, 1977.Google Scholar
Diebolt, J., and C. P., Robert. Estimation of finite mixture distributions through Bayesian sampling. Journal of the Royal Statistical Society, Series B, 56: 363–375, 1994.Google Scholar
Everitt, B. S., and D. J., Hand. Finite Mixture Distributions. Chapman and Hall, London, 1981.
Fleming, K. G., Yep, we're skewed. Variance, 2(2): 179–183, 2008.Google Scholar
Frees, E. W. Frequency and severity models. In E. W., Frees, G., Meyers, and R. A., Derrig (eds.), Predictive Modeling Applications in Actuarial Science: Volume 1, Predictive Modeling Techniques, pp. 138–164. Cambridge University Press, Cambridge, 2014.
Frees, E., P., Shi, and E., Valdez. Actuarial applications of a hierarchical claims model. ASTIN Bulletin, 39(1): 165–197, 2009.Google Scholar
Fu, L., and R., Moncher. Severity distributions for GLMs: Gamma or lognormal? Evidence from Monte Carlo simulations. Casualty Actuarial Society Discussion Paper Program, 2004.
Fu, L., and C. P., Wu. General iteration algorithms for classification ratemaking. Variance, 1(2): 193–213, 2007.Google Scholar
Gordon, S. K., and B., Jørgensen. Fitting Tweedie's compound Poisson model to insurance claims data: Dispersion modelling. ASTIN Bulletin, 32: 143–157, 2002.Google Scholar
Guven, S. Multivariate spatial analysis of the territory rating variable. Casualty Actuarial Society Discussion Paper Program, 2004.
Henry, J. B., and P., Hsieh. Extreme value analysis for partitioned insurance losses. Variance, 3(2): 214–238, 2009.Google Scholar
Holler, K., D., Sommer, and G., Trahair. Something old, something new in classification ratemaking with a novel use of generalized linear models for credit insurance. CAS Forum, Winter 1999.
Johnson, W. G., M. L., Baldwin, and R. J., Butler. The costs and outcomes of chiropractic and physician care for workers' compensation back claims. The Journal of Risk and Insurance, 66(2): 185–205, 1999.Google Scholar
Keatinge, C. Modeling losses with the mixed exponential distribution. Proceedings of the Casualty Actuarial Society, LXXXVI:654–698, 1999.Google Scholar
Kessler, D., and A., McDowell. Introducing the FMM procedure for finite mixture models. SAS Global Forum, 2012.
Klugman, S. A., H. E., Panjer, and G. E., Willmot. Loss Models: From Data to Decisions. John Wiley, New York, 1998.
Kremer, E. IBNR claims and the two way model of ANOVA. Scandinavian Actuarial Journal, 1982: 47–55, 1982.Google Scholar
Matthys, G., E., Delafosse, A., Guillou, and J., Beirlant. Estimating catastrophic quantile levels for heavy-tailed distributions. Insurance: Mathematics and Economics, 34: 517–537, 2004.Google Scholar
McCullagh, P., and J., Nelder. Generalized Linear Models. 2nd ed. Chapman and Hall/CRC, Boca Raton, FL, 1989.
McLachlan, G. J., and D., Peel. Finite Mixture Models. New York: John Wiley, 2000.
Miccolis, R. S. On the theory of increased limits and excess of loss pricing. Proceedings of Casualty Actuarial Society, LXIV:27–59, 1977.Google Scholar
Mildenhall, S. J. A systematic relationship between minimum bias and generalized linear models. Proceedings of the Casualty Actuarial Society, LXXXVI:393–487, 1999.Google Scholar
Mildenhall, S. J. Discussion of “general minimum bias models.” Casualty Actuarial Society Forum, Winter 2005.
Mosley, R. Estimating claim settlement values using GLM. Casualty Actuarial Society Discussion Paper Program, 2004.
Nelder, J. A., and O., Pregibon. An extended quasi-likelihood function. Biometrik, 74: 221–231, 1987.Google Scholar
Newcomb, S. A generalized theory of the combination of observations so as to obtain the best result. American Journal of Mathematics, 8: 343–366, 1886.Google Scholar
Parsa, R. A., and S. A., Klugman. Copula regression. Variance, 5(1): 45–54, 2011.Google Scholar
Pearson, K. Contributions to mathematical theory of evolution. Philosophical Transactions, Series A, 185: 71–110, 1894.Google Scholar
Pregibon, O. Review of generalized linear models. Ann. Statistics, 12: 1589–1596, 1984.Google Scholar
Renshaw, A. E. Chain ladder and interactive modelling (claims reserving and GLIM). Journal of the Institute of Actuaries, 116(3): 559–587, 1989.Google Scholar
Renshaw, A. E., and R. J., Verrall. A stochastic model underlying the chain ladder technique. Proceedings XXV ASTIN Colloquium, 1994.
Shi, P. Fat-tailed regression models. In E. W., Frees, G., Meyers, and R. A., Derrig (eds.), Predictive Modeling Applications in Actuarial Science: Volume 1, Predictive Modeling Techniques, pp. 236–259. Cambridge University Press, Cambridge, 2014.
Smyth, G. K. Generalized linear models with varying dispersion. J. R. Statist. Soc., Series B: 51:57–60, 1989.Google Scholar
Smyth, G. K., and B., Jørgensen. Fitting Tweedie's compound Poisson model to insurance claims data: Dispersion modelling. ASTIN Bulletin, 32: 143–157, 2002.Google Scholar
Smyth, G. K., and A. P., Verbyla. Adjusted likelihood methods for modelling dispersion ingeneralized linear models. Environmetrics, 10: 696–709, 1999.Google Scholar
Taylor, G. C. The chain ladder and Tweedie distributed claims data. Variance, 3(1): 96–104, 2009.Google Scholar
Taylor, G. C., and F. R., Ashe. Second moments of estimates of outstanding claims. Journal of Econometrics, 23: 37–61, 1983.Google Scholar
Teng, M. S. Pricing workers' compensation large deductible and excess insurance. Casualty Actuarial Society Forum, Winter 1994.
Titterington, D., A., Smith, and U., Makov. Statistical Analysis of Finite Mixture Distributions. John Wiley, New York, 1985.
Venter, G. Generalized linear models beyond the exponential family with loss reserve applications. Casualty Actuarial Society E-Forum, Summer 2007.
Verbyla, A. P., and G. K., Smyth. Double generalized linear models: Approximate residualmaximum likelihood and diagnostics. Research Report, Department of Statistics, University of Adelaide, 1998.
Wang, S. A set of new methods and tools for enterprise risk capital management and portfolio optimization. Casualty Actuarial Society Forum, Spring 2002.
Wedderburn, R. W. M. Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method. Biometrika, 61(3): 439–447, 1974.Google Scholar
Wedel, M., and W. S., DeSarbo. A mixture likelihood approach for generalized linear models. Journal of Classification, 12: 21–55, 1995.Google Scholar
Yan, J., J., Guszcza, M., Flynn, and C. P., Wu. Applications of the offset in property-casualty predictive modeling. Casualty Actuarial Society Forum, Winter 2009.

Save book to Kindle

To save this book to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats
×

Save book to Dropbox

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Dropbox.

Available formats
×

Save book to Google Drive

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Google Drive.

Available formats
×