Integrasi Extreme Value Theory Dan Value At Risk Dalam Pemodelan Harga Catastrophe Bond Banjir Berbasis Curah Hujan Di DKI Jakarta

Fathin, Laras (2026) Integrasi Extreme Value Theory Dan Value At Risk Dalam Pemodelan Harga Catastrophe Bond Banjir Berbasis Curah Hujan Di DKI Jakarta. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006221104-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Indonesia, khususnya DKI Jakarta, menghadapi risiko banjir yang tinggi akibat curah hujan ekstrem, sehingga diperlukan mekanisme pembiayaan risiko bencana yang cepat dan terukur. Catastrophe Bond (CAT Bond) merupakan salah satu instrumen insurance-linked securities yang memungkinkan transfer risiko finansial dari sponsor kepada investor di pasar modal. Penelitian ini bertujuan memodelkan curah hujan ekstrem sebagai trigger CAT Bond, mengkuantifikasi risiko kerugian menggunakan Value at Risk (VaR) dan Tail Value at Risk (TVaR), serta menentukan struktur dan harga wajar CAT Bond banjir bagi Pemerintah Provinsi DKI Jakarta. Curah hujan ekstrem dimodelkan menggunakan Extreme Value Theory (EVT) dengan metode Peaks Over Threshold (POT) dan distribusi Generalized Pareto (GPD). Penentuan threshold dilakukan melalui kombinasi Mean Excess Plot dan parameter stability plot, yang menghasilkan threshold pada kuantil 89% (u = 42,50 mm) dengan 160 data exceedance. Estimasi parameter menggunakan Maximum Likelihood Estimation melalui metode Newton–Raphson menghasilkan parameter shape ξ̂ = 0,160034 dan scale β̂ = 24,791869, dengan kecocokan model dikonfirmasi melalui uji Kolmogorov–Smirnov (p-value = 0,918943). Pada tingkat kepercayaan 99%, diperoleh VaR₀,₉₉ = 114,8541 mm sebagai attachment point dan TVaR₀,₉₉ = 158,1546 mm sebagai exhaustion point. Harga CAT Bond diestimasi menggunakan simulasi Monte Carlo dengan jumlah simulasi (M = 10.000) dan jumlah replikasi (K = 1.000) yang ditentukan berdasarkan target presisi standar galat (Glasserman, 2004). Hasil simulasi menunjukkan probabilitas No Loss sebesar 90,94%, Risk Layer sebesar 5,91%, dan Total Loss sebesar 3,15%, dengan estimasi harga wajar CAT Bond sebesar Rp10.162.235.083,89, yang dikonfirmasi stabil melalui rata-rata hasil replikasi sebesar Rp10.162.574.296,59. Nilai harga yang lebih tinggi dari nilai nominal obligasi (Rp10.000.000.000) mengindikasikan bahwa manfaat kupon secara ekspektasi melebihi ekspektasi kerugian akibat risiko banjir ekstrem. Hasil penelitian ini diharapkan dapat menjadi kerangka kuantitatif bagi pemerintah daerah dalam merancang instrumen pembiayaan risiko bencana berbasis pasar.
==================================================================================================================================
Indonesia, particularly the Special Capital Region of Jakarta (DKI Jakarta), faces a high risk of flooding caused by extreme rainfall, necessitating a fast and measurable disaster risk financing mechanism. A Catastrophe Bond (CAT Bond) is an insurance-linked securities instrument that enables the transfer of financial risk from a sponsor to investors in the capital market. This study aims to model extreme rainfall as a CAT Bond trigger, quantify loss risk using Value at Risk (VaR) and Tail Value at Risk (TVaR), and determine the structure and fair price of a flood CAT Bond for the DKI Jakarta Provincial Government. Extreme rainfall was modeled using Extreme Value Theory (EVT) with the Peaks Over Threshold (POT) method and the Generalized Pareto Distribution (GPD). Threshold selection was carried out through a combination of the Mean Excess Plot and the parameter stability plot, yielding a threshold at the 89th percentile (u = 42.50 mm) with 160 exceedance data points. Parameter estimation using Maximum Likelihood Estimation via the Newton–Raphson method produced a shape parameter of ξ̂ = 0.160034 and a scale parameter of β̂ = 24.791869, with goodness-of-fit confirmed through the Kolmogorov–Smirnov test (p-value = 0.918943). At the 99% confidence level, VaR₀.₉₉ = 114.8541 mm was obtained as the attachment point and TVaR₀.₉₉ = 158.1546 mm as the exhaustion point. The CAT Bond price was estimated using Monte Carlo simulation, with the number of simulations (M = 10,000) and the number of replications (K = 1,000) determined based on a target standard error precision (Glasserman, 2004). The simulation results yielded a No Loss probability of 90.94%, a Risk Layer probability of 5.91%, and a Total Loss probability of 3.15%, with an estimated fair CAT Bond price of Rp10,162,235,083.89, confirmed to be stable through a replication average of Rp10,162,574,296.59. The price being higher than the bond's face value (Rp10,000,000,000) indicates that the expected coupon benefit exceeds the expected loss arising from extreme flood risk. The findings of this study are expected to serve as a quantitative framework for local governments in designing market-based disaster risk financing instruments.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Catastrophe Bond, Extreme Value Theory, Generalized Pareto Distribution, Value at Risk, simulasi Monte Carlo, curah hujan ekstrem.
Subjects:	Q Science Q Science > QA Mathematics Q Science > QA Mathematics > QA614.58 Catastrophes
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Laras Inas Fathin
Date Deposited:	11 Aug 2026 09:34
Last Modified:	11 Aug 2026 09:34
URI:	http://repository.its.ac.id/id/eprint/144315

Actions (login required)

View Item