ESTIMASI CADANGAN KLAIM MENGGUNAKAN METODE KALMAN FILTER DENGAN STATE SPACE MODEL SCALAR PADA PRODUK ASURANSI UMUM


Chintya Carissa Manurung(1*), Tiara Yulita(2), Amalia Listiani(3)

(1) Sumatera Institute of Technology, Indonesia
(2) Institut Teknologi Sumatera, Indonesia
(3) Institut Teknologi Sumatera, Indonesia
(*) Corresponding Author

Abstract


Usually, there is a delay in reporting claims from the time of the incident which results in the insurance company having a responsibility or debt. Therefore, insurance companies need to prepare funds to cover these debts, namely with claims reserves. There are two types of claim reserves, namely Incurred But Not Reported (IBNR) and Reported But Not Settled (RBNS). This research focuses on determining the estimation of aggregate claim reserves using the Kalman Filter method with scalar State Space Models (SSMs) which is a model resulting from the development of the Chain Ladder (CL) method. The Kalman Filter method with SSMs is a stochastic method that takes into account the time series model so that it can predict the temporal dynamics of a system more accurately. The results of forecasting claim reserves using the Kalman Filter method with SSMs will be compared with the CL method. Variational Of Coefficient (VOC) is an error predictor to determine the best method. The calculation results using the Kalman Filter method with SSMs produce a smaller VOC value than the CL method, proving that the Kalman Filter method with SSMs is better than CL.

Keywords


Asuransi Umum, Cadangan Klaim, Kalman Filter , State Space Model Scalar, Chain Ladder

References


K. S. Sundanis Agung Pertiwi1, I Nyoman Widana, “Estimasi Cadangan Klaim pada Asuransi Umum dengan Metode Chain Ladder,” E-Jurnal Mat., vol. 12, 2023.

Y. Hikmah and I. R. Hikmah, “Perhitungan Cadangan Klaim Dengan Metode Chain Ladder Menggunakan Excel Dan Rstudio,” MAp (Mathematics Appl. J., vol. 4, no. 2, pp. 122–131, 2022, doi: 10.15548/map.v4i2.4837.

L. Bayu Alamsyah Pratama, Perhitungan Cadangan Klaim dengan Menggunkan Metode Chain-Ladder dan Munich Chain-Ladder, no. July. 2020.

N. Chukhrova and A. Johannssen, “Kalman filter learning algorithms and state space representations for stochastic claims reserving,” Risks, vol. 9, no. 6, 2021, doi: 10.3390/risks9060112.

N. Chukhrova and A. Johannssen, “State space models and the KALMAN-filter in stochastic claims reserving: Forecasting, filtering and smoothing,” Risks, vol. 5, no. 2, 2017, doi: 10.3390/risks5020030.

P. de Jong and B. Zehnwirth, “Claims reserving, state-space models and the Kalman filter,” J. Inst. Actuar., vol. 110, no. 01, pp. 157–181, 1983, doi: 10.1017/s0020268100041287.

T. Yulita and A. R. Effendie, “Estimation of Ibnr and Rbns Reserves Using Rdc Method and Gamma Generalized Linear Model,” Media Stat., vol. 15, no. 1, pp. 24–35, 2022, doi: 10.14710/medstat.15.1.24-35.

M. Elok Nurul Islam, Y. Wilandari, M. Jurusan Statistika FSM Universitas Diponegoro, and S. Pengajar Jurusan Statistika, “Perhitungan Pembiayaan Dana Pensiun Dengan Metode Attained Age Normal Dan Projected Unit Credit (Studi Kasus : Pt. Taspen (Persero) Kantor Cabang Utama Semarang),” J. Gaussian, vol. 5, no. 3, pp. 505–514, 2016, [Online]. Available: http://ejournal-s1.undip.ac.id/index.php/gaussian

B. Y. T. Mack, “Distribution-Free Calculation Of The Standard Error Of Chain Ladder Reserve Estimates,” vol. 23, no. 2, 1993.

A. K. Mutaqin, D. R. Tampubolon, and S. Darwis, “Run-Off Triangle Data dan Permasalahannya,” vol. 8, no. 1, pp. 55–59, 2008.

C. Manurung, “Estimasi cadangan klaim menggunakan metode kalman filter dengan state space model scalar pada produk asuransi umum,” 2024.


Refbacks

  • There are currently no refbacks.


Copyright (c) 2024 Chintya Carissa Manurung, Tiara Yulita S.Pd., M.Sc., Amalia Listiani, S.Pd., M.Sc.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.


Journal KUBIK: Jurnal Publikasi Ilmiah Matematika has indexed by:

SINTA DOAJ Dimensions Google Scholar Garuda Moraref DOI Crossref

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.