×

A non-convex regularization approach for stable estimation of loss development factors. (English) Zbl 1479.91329

The paper focuses on non-convex regularization methods aimed at obtaining stable estimation of loss development factors in insurance claims reserving. After presenting the peculiarities and the use of the log-adjusted absolute deviation (LAAD) penalty, some issues on optimization of LAAD penalized regression model are in-depth. Then, a simulation study highlights the efficacy of the proposed method. Empirical applications show the possibility to use of LAAD penalty in insurance claims reserving, considering an insurer’s dataset with multi-line reported loss triangles. A comparison to other regression models points out the efficacy of the LAAD penalized regression model.

MSC:

91G05 Actuarial mathematics
PDFBibTeX XMLCite
Full Text: DOI arXiv

References:

[1] Armagan, A.; Dunson, D. B.; Lee, J., Generalized double Pareto shrinkage, Statistica Sinica, 23, 1, 119-143 (2013) · Zbl 1259.62061
[2] Bruckstein, A. M.; Donoho, D. L.; Elad, M., From sparse solutions of systems of equations to sparse modeling of signals and images, SIAM Review, 51, 1, 34-81 (2009) · Zbl 1178.68619 · doi:10.1137/060657704
[3] Chan, F.; Chan, L.; Mead, E., Properties and modifications of Whittaker-Henderson graduation, Scandinavian Actuarial Journal, 1982, 57-61 (1982) · Zbl 0502.62089 · doi:10.1080/03461238.1982.10405433
[4] Devriendt, S.; Antonio, K.; Reynkens, T.; Verbelen, R., Sparse regression with multi-type regularized feature modeling, Insurance: Mathematics and Economics, 96, 248-261 (2021) · Zbl 1460.91218
[5] Efron, B., How biased is the apparent error rate of a prediction rule?, Journal of the American Statistical Association, 81, 394, 461-470 (1986) · Zbl 0621.62073 · doi:10.1080/01621459.1986.10478291
[6] Fan, J.; Li, R., Variable selection via nonconcave penalized likelihood and its oracle properties, Journal of the American Statistical Association, 96, 456, 1348-1360 (2001) · Zbl 1073.62547 · doi:10.1198/016214501753382273
[7] Frees, E. W.; Lee, G.; Yang, L., Multivariate frequency-severity regression models in insurance, Risks, 4, 1, 4 (2016) · doi:10.3390/risks4010004
[8] Friedman, J., Hastie, T. & Tibshirani, R. (2009). glmnet: Lasso and elastic-net regularized generalized linear models. R package version 4.0-2.
[9] Gao, G., Bayesian claims reserving methods in non-life insurance with STAN (2018), Singapore: Springer, Singapore
[10] Hastie, T.; Tibshirani, R.; Friedman, J., The elements of statistical learning: data mining, inference, and prediction (2009), Berlin: Springer Science & Business Media, Berlin · Zbl 1273.62005
[11] Hastie, T.; Tibshirani, R.; Wainwright, M., Statistical learning with sparsity: the lasso and generalizations (2015), Boca Raton, Florida: CRC Press, Boca Raton, Florida · Zbl 1319.68003
[12] Hoerl, A. E.; Kennard, R. W., Ridge regression: biased estimation for nonorthogonal problems, Technometrics, 12, 1, 55-67 (1970) · Zbl 0202.17205 · doi:10.1080/00401706.1970.10488634
[13] James, G.; Witten, D.; Hastie, T.; Tibshirani, R., An introduction to statistical learning (2013), New York, New York: Springer, New York, New York · Zbl 1281.62147
[14] Jeong, H., Testing for random effects in compound risk models via Bregman divergence, ASTIN Bulletin: The Journal of the IAA, 50, 3, 777-798 (2020) · Zbl 1454.91194 · doi:10.1017/asb.2020.19
[15] Jeong, H.; Dey, D. K., Application of vine copula for multi-line insurance reserving, Risks, 8, 4, 111 (2020) · doi:10.3390/risks8040111
[16] Knight, K.; Fu, W., Asymptotics for lasso-type estimators, The Annals of Statistics, 28, 5, 1356-1378 (2000) · Zbl 1105.62357 · doi:10.1214/aos/1015957397
[17] Kyung, M.; Gill, J.; Ghosh, M.; Casella, G., Penalized regression, standard errors, and Bayesian lassos, Bayesian Analysis, 5, 2, 369-411 (2010) · Zbl 1330.62289 · doi:10.1214/10-BA607
[18] Lee, A., Caron, F., Doucet, A. & Holmes, C. (2010). A hierarchical Bayesian framework for constructing sparsity-inducing priors. arXiv preprint arXiv:1009.1914.
[19] Lee, G. Y.; Shi, P., A dependent frequency-severity approach to modeling longitudinal insurance claims, Insurance: Mathematics and Economics, 87, 115-129 (2019) · Zbl 1410.91271
[20] Lumley, T. (2013). R package ‘leaps’. R package version 3.1.
[21] Luo, Z.-Q.; Tseng, P., On the convergence of the coordinate descent method for convex differentiable minimization, Journal of Optimization Theory and Applications, 72, 1, 7-35 (1992) · Zbl 0795.90069 · doi:10.1007/BF00939948
[22] Mack, T., Distribution-free calculation of the standard error of chain ladder reserve estimates, ASTIN Bulletin: The Journal of the IAA, 23, 2, 213-225 (1993) · doi:10.2143/AST.23.2.2005092
[23] Mack, T., The standard error of chain ladder reserve estimates: recursive calculation and inclusion of a tail factor, ASTIN Bulletin: The Journal of the IAA, 29, 2, 361-366 (1999) · Zbl 1277.62256 · doi:10.2143/AST.29.2.504622
[24] Mazumder, R.; Friedman, J. H.; Hastie, T., Sparsenet: coordinate descent with nonconvex penalties, Journal of the American Statistical Association, 106, 495, 1125-1138 (2011) · Zbl 1229.62091 · doi:10.1198/jasa.2011.tm09738
[25] McGuire, G., Taylor, G. & Miller, H. (2018). Self-assembling insurance claim models using regularized regression and machine learning. Available at SSRN 3241906
[26] Nawar, S. M. (2016). Machine learning techniques for detecting hierarchical interactions in insurance claims models. Master’s thesis, Concordia University.
[27] Park, T.; Casella, G., The Bayesian LASSO, Journal of the American Statistical Association, 103, 482, 681-686 (2008) · Zbl 1330.62292 · doi:10.1198/016214508000000337
[28] Renshaw, A. E., Chain ladder and interactive modelling (claims reserving and glim), Journal of the Institute of Actuaries, 116, 3, 559-587 (1989) · doi:10.1017/S0020268100036702
[29] Shi, P.; Basu, S.; Meyers, G. G., A Bayesian log-normal model for multivariate loss reserving, North American Actuarial Journal, 16, 1, 29-51 (2012) · Zbl 1291.91126 · doi:10.1080/10920277.2012.10590631
[30] Shi, P.; Frees, E. W., Dependent loss reserving using copulas, ASTIN Bulletin: The Journal of the IAA, 41, 2, 449-486 (2011)
[31] Stein, C. M., Estimation of the mean of a multivariate normal distribution, The Annals of Statistics, 9, 6, 1135-1151 (1981) · Zbl 0476.62035 · doi:10.1214/aos/1176345632
[32] Taylor, G.; McGuire, G., Stochastic loss reserving using generalized linear models (2016), Arlington, Virginia: Casualty Actuarial Society, Arlington, Virginia
[33] Tibshirani, R., Regression shrinkage and selection via the LASSO, Journal of the Royal Statistical Society. Series B (Methodological), 58, 1, 267-288 (1996) · Zbl 0850.62538 · doi:10.1111/j.2517-6161.1996.tb02080.x
[34] Tseng, P., Convergence of a block coordinate descent method for nondifferentiable minimization, Journal of Optimization Theory and Applications, 109, 3, 475-494 (2001) · Zbl 1006.65062 · doi:10.1023/A:1017501703105
[35] Wang, S.; Zhang, H.; Chai, H.; Liang, Y., A novel log penalty in a path seeking scheme for biomarker selection, Technology and Health Care, 27, S1, 85-93 (2019) · doi:10.3233/THC-199009
[36] Williams, B., Hansen, G., Baraban, A. & Santoni, A. (2015). A practical approach to variable selection – a comparison of various techniques. In Casualty Actuarial Society E-Forum, P. 4-40. Arlington, Virginia: Casualty Actuarial Society.
[37] Yin, C.; Lin, X. S., Efficient estimation of erlang mixtures using iscad penalty with insurance application, ASTIN Bulletin: The Journal of the IAA, 46, 3, 779-799 (2016) · Zbl 1390.62030 · doi:10.1017/asb.2016.14
[38] Zhang, C.-H., Nearly unbiased variable selection under minimax concave penalty, The Annals of Statistics, 38, 2, 894-942 (2010) · Zbl 1183.62120 · doi:10.1214/09-AOS729
[39] Zhang, C.-H. & Melnik, O. (2009). plus: Penalized linear unbiased selection. R package version 1.0.
[40] Zou, H.; Hastie, T.; Tibshirani, R., On the ‘degrees of freedom’ of the lasso, The Annals of Statistics, 35, 5, 2173-2192 (2007) · Zbl 1126.62061 · doi:10.1214/009053607000000127
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.