Prasetio, Akmal (2026) Aplikasi Vine Copula Dalam Estimasi Liabilitas Dan Penyesuaian Risiko Untuk Portofolio Multi-Lini Bisnis Berdasarkan IFRS 17. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5006221056_Undergraduate_thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5006221056_Undergraduate_thesis.pdf Restricted to Repository staff only until 14 January 2026. Download (5MB) | Request a copy |
Abstract
Sektor asuransi memegang peran penting dalam pengelolaan risiko dengan memberikan perlindungan kepada pemegang polis, namun efektivitasnya sangat bergantung pada kemampuan perusahaan dalam memenuhi kewajiban klaim. International Financial Reporting Standard 17 (IFRS 17) menekankan pentingnya pengukuran kewajiban kontrak asuransi berbasis fulfilment cash flows, termasuk best estimate liability (BEL) dan risk adjustment (RA) yang mencerminkan ketidakpastian risiko onkeuangan, sehingga estimasi liability for incurred claims (LIC) harus mempertimbangkan distribusi penuh unpaid claims. Pendekatan konvensional sering mengabaikan ketergantungan antar lini bisnis, sehingga penelitian ini menerapkan vine copula untuk memodelkan hubungan antar lima lini bisnis utama, yaitu Motor, Personal Accident & Workers’ Compensation, Property, Credit, dan Marine, menggunakan data cumulative paid loss triangles Munich Re periode 2013–024. Distribusi marginal tiap lini diestimasi melalui pendekatan Bayesian dengan MCMC, dan hasil pemilihan distribusi marginal menunjukkan lognormal constrained sebagai model terbaik. Struktur dependensi terbaik yang terpilih adalah regular vine, yang memungkinkan simulasi Monte Carlo untuk memperoleh distribusi unpaid losses baik pada tingkat lini bisnis maupun portofolio, dengan mempertimbangkan hubungan yang saling terkait. Hasil penelitian menunjukkan nilai BEL (dalam Euro) sebesar 43.376.062.898 pada tingkat portofolio. Sedangkan RA pada tingkat portofolio dan lini bisnis dihitung berdasarkan berbagai ukuran risiko, yakni VaR pada level konfidensi 75% sebesar 10.802.290.678, CTE pada level konfidensi 5% sebesar 26.031.183.359, Wang Transform sebesar 30.599.207.337 pada preferensi risiko 10%, 42.857.696.850 pada preferensi risiko 5%, dan 72.687.694.174 pada preferensi risiko 1%, serta Cost of Capital (CoC) sebesar 5.744.583.608 pada level konfidensi 99%, 3.370.794.522 pada level konfidensi 95%, dan 5.379.695.588 pada level konfidensi 90%. LIC yang diperoleh dari penjumlahan BEL dan RA memiliki nilai 54.178.353.576 untuk RA VaR 75%, 69.407.246.258 untuk RA CTE 75%, 73.975.270.235 untuk RA Wang Transform 10%, 86.233.759.748 untuk RA Wang Transform 5%, 116.063.757.072 untuk RA Wang Transform 1%, serta 48.755.758.487 untuk RA CoC 99%, 46.746.857.420 untuk RA CoC 95%, dan 45.907.985.322 untuk RA CoC 90%.
=============================================================================================================================
The insurance sector plays a crucial role in risk management by providing protection to policyholders, yet its effectiveness largely depends on the company’s ability to meet claim obligations. International Financial Reporting Standard 17 (IFRS 17) emphasizes the measurement of insurance contract liabilities based on fulfilment cash flows, including best estimate liability (BEL) and risk adjustment (RA), which reflect non-financial risk uncertainty, thus the estimation of liability for incurred claims (LIC) must consider the full distribution of unpaid claims. Conventional approaches often overlook dependencies across business lines, so this study applies vine copula to model the relationships among five main business lines, namely Motor, Personal Accident & Workers’ Compensation, Property, Credit, and Marine, using cumulative paid loss triangles from Munich Re for the period 2013–2024. The marginal distributions f each line were estimated using a Bayesian approach via MCMC, and the selection results indicate that the lognormal constrained distribution is the best fit. The optimal dependence structure selected is a regular vine, which allows Monte Carlo simulation to obtain the distribution of unpaid losses at both the business line and portfolio levels, taking into account the interrelated development factors. The study results show a BEL of 43,376,062,898 at the portfolio level. Meanwhile, RA at both portfolio and business line levels is calculated based on various risk measures, namely VaR at 75% confidence level of 10,802,290,678, CTE at 75% confidence level of 26,031,183,359, Wang Transform of 30,599,207,337 at 10% risk preference, 42,857,696,850 at 5% risk preference, and 72,687,694,174 at 1% risk preference, as well as Cost of Capital (CoC) of 5,744,583,608 at 99% confidence level, 3,370,794,522 at 95%, and 5,379,695,588 at 90%. LIC obtained by summing BEL and RA amounts to 54,178,353,576 for RA VaR 75%, 69,407,246,258 for RA CTE 75%, 73,975,270,235 for RA Wang Transform 10%, 86,233,759,748 for RA Wang Transform 5%, 116,063,757,072 for RA Wang Transform 1%, and 48,755,758,487 for RA CoC 99%, 46,746,857,420 for RA CoC 95%, and 45,907,985,322 for RA CoC 90%.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Development Factor, Markov Chain Monte Carlo, Liability for Incurred Claims, Risk Adjustment, Vine Copula. |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HG Finance H Social Sciences > HG Finance > HG4012 Mathematical models H Social Sciences > HG Finance > HG4028.V3 Valuation. Economic value H Social Sciences > HG Finance > HG8051 Insurance H Social Sciences > HG Finance > HG8054.5 Risk (Insurance) Q Science > QA Mathematics > QA401 Mathematical models. |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis |
| Depositing User: | Akmal Adi Prasetio |
| Date Deposited: | 13 Jan 2026 01:23 |
| Last Modified: | 13 Jan 2026 01:23 |
| URI: | http://repository.its.ac.id/id/eprint/129524 |
Actions (login required)
 |
View Item |