Document Type : Research Article
Authors
1 Department of Actuarial Science, Shahid Beheshti University, Theran, Iran
2 Department of Actuarial Science, Shahid Beheshti University, Tehran, Iran
Abstract
An important question in non life insurance research is the estimation of number of future payments and corresponding amount of them. A loss reserve is the money set aside by insurance companies to pay policyholders claims on their policies. The policyholder behavior for reporting claims after its occurrence have significant effect on the costs of the insurance company. This article considers the problem of predicting the amount and number of claims that have been incurred but not reported, say IBNR. Using the delay probabilities in monthly level, calculated by the Zero Inflated Gamma Mixture distribution, it predicts IBNR's loss reserve. The model advantage in the IBNR reserve is insurers can predict the number of future claims for each future date. This enables them to change the claim reporting process. The practical applications of our findings are applied against a third party liability (TPL) insurance loss portfolio. Additional information about claim can be considered in the loss reserving model and making the prediction of amount more accurate.
Keywords
[2] A.L. BADESCU, X.S. LIN, D. TANG, A marked Cox model for the number of IBNR claims: theory, Insurance: Mathematics and Economics, 69(2016), 29-37.
[3] A.L. BADESCU, T. CHEN, X.S. LIN, AND D. TANG, A marked COX model for the number of IBNR claims: estimation and application, ASTIN Bulletin: The Journal of the IAA, 49(2019), 709-739.
[4] J. CREVECOEUR, K. ANTONIO, AND R. VERBELEN, A Time Change Strategy to Model Reporting Delay Dynamics in Claims Reserving, The 3rd Europen Actuarial Journal conference, (2018), KU Leuven, Belgium.
[5] J. CREVECOEUR, K. ANTONIO, AND R. VERBELEN, Modeling the number of hidden events subject to observation delay, European Journal of Operational Research, 277(2019), 930-944.
[6] V.R. DA SILVA, A. YONGACOGLU, EM algorithm on the approximation of arbitrary pdfs by gaussian, gamma and lognormal mixture distributions, The 7th IEEE Latin-American Conference on Communications, LATINCOM, IEEE,
(2015).
[7] M. GHARIB, Two characterisations of a gamma mixture distribution, Bulletin of the Australian Mathematical Society, 52(1995), 353-358.
[8] M.D. MARTINEZ-MIRANDA, J.P. NIELSEN, AND R.J. VERRALL, Double chain ladder, ASTIN Bulletin: The Journal of the IAA, 42(2012), 59-76.
[9] M.D. MARTINEZ-MIRANDA, J.P. NIELSEN AND R.J. VERRALL, Double chain ladder and Bornhuetter-Ferguson, North American Actuarial Journal, 17(2013a), 101-113.
[10] M.D. MARTINEZ-MIRANDA, J.P. NIELSEN, S. SPERLICH, AND R.J. VERRALL, Continuous Chain Ladder: Reformulating and generalizing a classical insurance problem, Expert Systems with Applications, 40(2013b), 5588-5603.
[12] R. VERBELEN, K. ANTONIO, G. CLAESKENS, AND J. CREVECOEUR, Modeling the occurrence of events subject to a reporting delay via an EM algorithm, arXiv:1909.08336, (2019).
[13] R.J. VERRALL, J.P. NIELSEN, AND A. H. JESSEN, Prediction of RBNS and IBNR Claims using Claim Amounts and Claim Counts, ASTIN Bulletin: The Journal of the IAA, 40(2010), 871-887.
[14] R.J. VERRALL, M.V. WUTHRICH, Understanding reporting delay in general insurance, Risks, 4(2016),1-36.
)