VAR pristup: Creditmetrics

master's thesis

• Author: Ana Bokšić
• Mentor: Boris Podobnik (mentor)
• Committee member: Boris Podobnik (predsjednik povjerenstva)
• Committee member: Snježana Lubura Strunjak (član povjerenstva)
• Committee member: Igor Ciganović (član povjerenstva)
• Committee member: Igor Pažanin (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2019-09-25, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

Otkako je Banka za međunarodna poravnavanja definirala zahtjeve na kapitalne rezerve VaR metodologija postajala je sve važnija u bankovnoj praksi. Bankama je dozvoljeno koristiti vlastite interne modele za računanje izloženosti kreditnom riziku. Prvo uvodimo osnovne koncepte VaR-a, a onda gledamo njegovo potencijalno proširenje na neutržive kredite, te njegovu ulogu u računanju kapitalnih pričuva banke. Posebnu pažnju poklanjamo CreditMetrics-u originalno predloženom od strane J.P. Morgana. CreditMetrics predstavlja korisno pomagalo u VaR modeliranju kreditiranja. Primjenjujemo koncepte kao što su Generalizirana Pareto distribucija s teškim repovima i procjena neočekivanih gubitaka primjenom teorije ekstremne vrijednosti. Predstavljamo metodologiju modela koji koristi Monte Carlo simulaciju za procjenu različitih mjera kreditnog rizika. Na kraju iznosimo rezultate simulacije za hipotetski portfelj obveznica.

Abstract (english)

Since the Bank for International Settlements introduced capital requirements for market risk, VAR methodologies have become more and more important in banking. Certain banks were allowed to develop and use internal VAR models for calculating the exposure to credit risk. Firstly, we review basic VAR concept and then look at its potential extension to nontradable loans and its use in calculating the bank's capital requirement. Special attention is paid to J.P.Morgan's CreditMetrics, which is a useful tool in VAR modelling for credit risk. We apply concepts such as Generalized Pareto distribution with heavy tails and estimate unexpected loss using extreme value theory. We introduce the methodology for Monte Carlo simulation used to assess different credit risk measures. Lastly, we carry out the results of the simulation for a hypothetical loan portfolio.

• Keywords: VaR metodologija; kreditni rizik; CreditMetrics; Monte Carlo simulacija
• Keywords (english): VAR methodology; credit risk; CreditMetrics; Monte Carlo simulation
• Language: croatian
• URN:NBN: urn:nbn:hr:217:882890
• Study programme: Title: Finance and Business Mathematics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra matematike (magistar/magistra matematike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2020-01-13 10:52:51