Perbandingan Estimasi Cadangan Klaim Menggunakan Metode Bornhuetter-Ferguson, Bornhuetter-Double Chain Ladder, dan Generalized Linear Model

Iswara, Widya (2025) Perbandingan Estimasi Cadangan Klaim Menggunakan Metode Bornhuetter-Ferguson, Bornhuetter-Double Chain Ladder, dan Generalized Linear Model. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006211015-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (1MB) | Request a copy

Abstract

Semua aktivitas manusia dan perusahaan akan selalu berhadapan dengan risiko. Risiko merupakan ketidakpastian yang dapat menyebabkan kerugian. Salah satu cara untuk menanggulangi risiko adalah dengan menyediakan perlindungan melalui asuransi. Asuransi adalah bentuk pengalihan risiko dari tertanggung kepada pihak penanggung dengan mekanisme tertanggung membayar sejumlah premi kepada perusahaan asuransi. Ketika terjadi suatu kerugian, tertanggung dapat mengajukan klaim untuk memperoleh ganti rugi sesuai perjanjian dengan perusahaan asuransi. Perusahaan asuransi berkewajiban untuk membentuk cadangan klaim, yaitu dana yang disiapkan untuk membayar klaim yang belum terselesaikan. Perusahaan asuransi berupaya untuk memprediksi nilai cadangan klaim setepat mungkin dengan metode perhitungan terbaik agar dapat membayar kewajiban klaim kepada nasabah tanpa mengganggu operasional dan keuangan perusahaan. Beberapa metode yang umum digunakan untuk mengestimasi cadangan klaim meliputi metode deterministik dan stokastik. Metode Chain Ladder dan Bornhuetter-Ferguson (BF) merupakan metode deterministik yang paling umum digunakan. Pengembangan dari metode tersebut adalah Bornhuetter-Double Chain Ladder (BDCL) yang memiliki keunggulan dalam mengestimasi cadangan klaim IBNR dan RBNS secara terpisah. Metode Generalized Linear Model (GLM) yang merupakan metode stokastik dapat menutupi kekurangan dari metode deterministik dan memungkinkan dalam fleksibilitas pemilihan parameter. Penelitian ini membahas perbandingan hasil estimasi cadangan klaim yang diperoleh dengan metode Bornhuetter-Ferguson, Bornhuetter-Double Chain Ladder, dan Generalized Linear Model. Data klaim yang digunakan bersumber dari PT Reasuransi XYZ periode 2022 – 2023. Penentuan metode terbaik didasarkan pada tingkat akurasi menggunakan Mean Absolute Percentage Error (MAPE). Berdasarkan hasil MAPE, metode Bornhuetter-Ferguson dan Generalized Linear Model memberikan nilai yang lebih baik yaitu 15,2641% dibandingkan metode Bornhuetter-Double Chain Ladder. Total estimasi cadangan klaim berdasarkan metode Bornhuetter-Ferguson adalah Rp371.023.291.231, sedangkan dengan metode Generalized Linear Model adalah Rp371.023.291.396.
==================================================================================================================================
All human and corporate activities will always be faced with risk. Risk is uncertainty that can cause loss. One way to mitigate risk is to provide protection through insurance. Insurance is a form of risk transfer from the insured to the insurer with the mechanism of the insured paying a premium to the insurance company. When a loss occurs, the insured can file a claim to obtain compensation according to the agreement with the insurance company. Insurance companies are obliged to establish claim reserves, which are funds prepared to pay unresolved claims. Insurance companies try to predict the value of claim reserves as precisely as possible with the best calculation method in order to pay claim obligations to customers without disrupting the company's operations and finances. Some commonly used methods for estimating claim reserves include deterministic and stochastic methods. Chain Ladder and Bornhuetter-Ferguson (BF) methods are the most commonly used deterministic methods. The development of the method is Bornhuetter-Double Chain Ladder (BDCL) which has the advantage of estimating IBNR and RBNS claim reserves separately. The Generalized Linear Model (GLM) method which is a stochastic method can cover the shortcomings of deterministic methods and allows for flexibility in parameter selection. This research discusses the comparison of claim reserve estimation results obtained by the Bornhuetter-Ferguson, Bornhuetter-Double Chain Ladder, and Generalized Linear Model methods. The claim data used is sourced from PT Reasuransi XYZ for the period 2022 - 2023. Determination of the best method is based on the level of accuracy using Mean Absolute Percentage Error (MAPE). The results of this study are expected to provide information and other alternatives to insurance companies regarding the best methods that can be used to obtain estimates of claim reserves. Based on the results of the Mean Absolute Percentage Error (MAPE), the Bornhuetter-Ferguson and Generalized Linear Model methods exhibited better accuracy, achieving a value of 15.2641%, compared to the Bornhuetter-Double Chain Ladder. The total claim reserve estimated using the Bornhuetter-Ferguson method amounted to Rp371,023,291,231, while the Generalized Linear Model method is Rp371,023,291,396.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Bornhuetter-Ferguson, Bornhuetter-Double Chain Ladder, Claim Reserve, Generalized Linear Model, Cadangan Klaim
Subjects:	H Social Sciences > HG Finance > HG8771 Life insurance
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Widya Iswara
Date Deposited:	07 Jan 2025 07:30
Last Modified:	07 Jan 2025 07:30
URI:	http://repository.its.ac.id/id/eprint/116207

Actions (login required)

View Item