Perhitungan Capital Charge Untuk Risiko Operasional Perbankan : Metode Simulasi  

Oleh:

Lestari, Neni ; Hermanto, Bambang

Jenis: Article from Journal - ilmiah nasional - tidak terakreditasi DIKTI
Dalam koleksi: Manajemen Usahawan Indonesia vol. 35 no. 10 (Oct. 2006), page 16-27.
Topik: Simulasi; capital charge; risiko operasional

Ketersediaan

Perpustakaan Pusat (Semanggi) Nomor Panggil: MM15.27 Non-tandon: 1 (dapat dipinjam: 0) Tandon: tidak ada

Lihat Detail Induk

Isi artikel

The New Basel Accord II as an international regulation for banking industry will be effective at year end 2006 by basel commitee on banking supervision. This new basel will straighten up the capital charge for few significance risks in banking industry such as credit risk, liquidity risk and the new introduced risk, the operational risk. Since this new regulation will be effective soon and banking industry in indonesia is not familiar yet, the writer try to help them by designing the matrix and operational loss data processing procedure for capital charge calculation and causal modelling to deal with the regulation and to detect the factors which led to operational loss. The matrix is arrange from hourly to monthly event according to business lines and even types as in basel accord II. To handle the small amount of operational loss data, the writer has used bootstrap method. Bottom up approach is selected to modelling the operational risk. This approach using peaks over threshold generalized pareto distribution (POT-GDP) plus peaks over threshold point process (POT-PP) from extreme value theory to calculate the capital charge and also using the linear regression for causal modeling. The writer found that the larger the sample size is, the harder these distributions to fit with generalized pareto distribution in particular threshold especially for sample size which nearly 50 it can be assumed normal distribution so value at risk can be used to calculate the capital charge, even though the writer has tried to remove the threshold to the higher percentile. So it was suggested to the user of generalized pareto distribution always to fit their data or distributions using the kolmogrov smirnov statistics.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali