Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective.

Risk management is an integral factor facilitating global financial stability. Since 1988, the Basel Accord has been setting minimum regulatory capital standards for banks throughout the world based on individual bank’s Value-at-Risk (VaR) forecasts. Adopted by more than 100 countries, including Sin...

Full description

Saved in:
Bibliographic Details
Main Author: Ong, Yu Hui.
Other Authors: School of Humanities and Social Sciences
Format: Final Year Project
Language:English
Published: 2012
Subjects:
Online Access:http://hdl.handle.net/10356/49377
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-49377
record_format dspace
spelling sg-ntu-dr.10356-493772019-12-10T11:14:19Z Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective. Ong, Yu Hui. School of Humanities and Social Sciences Wang Wei-Siang DRNTU::Humanities Risk management is an integral factor facilitating global financial stability. Since 1988, the Basel Accord has been setting minimum regulatory capital standards for banks throughout the world based on individual bank’s Value-at-Risk (VaR) forecasts. Adopted by more than 100 countries, including Singapore, the current Basel Accord Market Risk framework is designed to encourage superior risk management by allowing banks to use internal models to calculate the VaR used in the determination of their daily capital requirement. Following the global financial turbulence in 2008-2009 and the debate on limitations of current VaR-based capital requirements, it becomes of paramount importance to search for VaR models which can overcome these limitations, to be used during the recent Europe Debt Crisis. This paper aims to evaluate the VaR forecasting performances of three GARCH specification models implemented on simple portfolios consisting of the Strait Times Index (STI), using five different sample sizes. Other than standard evaluation criteria such as Unconditional Coverage test and Quantile Loss Function, criterion based on minimization of Capital Requirements imposed by the Basel Accord was also considered. Test results showed that the optimal VaR model varies with sample size, nature of data used for estimation and different evaluation criteria. To prevent underestimation of VaR and inadequate capital, EGARCH models with a small sample size (i.e. with the most recent data) are desirable, but to balance against overestimation of VaR and unnecessary capital holding, TGARCH models with sufficiently large sample data are critical. Finally, results based on comparison between capital requirements and quantile losses also shed light on the advantages and limitations of the inclusion of Stressed-VaR in the determination of Basel III Accord Capital Requirement. Bachelor of Arts 2012-05-18T02:08:35Z 2012-05-18T02:08:35Z 2012 2012 Final Year Project (FYP) http://hdl.handle.net/10356/49377 en Nanyang Technological University 48 p. application/pdf
institution Nanyang Technological University
building NTU Library
country Singapore
collection DR-NTU
language English
topic DRNTU::Humanities
spellingShingle DRNTU::Humanities
Ong, Yu Hui.
Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective.
description Risk management is an integral factor facilitating global financial stability. Since 1988, the Basel Accord has been setting minimum regulatory capital standards for banks throughout the world based on individual bank’s Value-at-Risk (VaR) forecasts. Adopted by more than 100 countries, including Singapore, the current Basel Accord Market Risk framework is designed to encourage superior risk management by allowing banks to use internal models to calculate the VaR used in the determination of their daily capital requirement. Following the global financial turbulence in 2008-2009 and the debate on limitations of current VaR-based capital requirements, it becomes of paramount importance to search for VaR models which can overcome these limitations, to be used during the recent Europe Debt Crisis. This paper aims to evaluate the VaR forecasting performances of three GARCH specification models implemented on simple portfolios consisting of the Strait Times Index (STI), using five different sample sizes. Other than standard evaluation criteria such as Unconditional Coverage test and Quantile Loss Function, criterion based on minimization of Capital Requirements imposed by the Basel Accord was also considered. Test results showed that the optimal VaR model varies with sample size, nature of data used for estimation and different evaluation criteria. To prevent underestimation of VaR and inadequate capital, EGARCH models with a small sample size (i.e. with the most recent data) are desirable, but to balance against overestimation of VaR and unnecessary capital holding, TGARCH models with sufficiently large sample data are critical. Finally, results based on comparison between capital requirements and quantile losses also shed light on the advantages and limitations of the inclusion of Stressed-VaR in the determination of Basel III Accord Capital Requirement.
author2 School of Humanities and Social Sciences
author_facet School of Humanities and Social Sciences
Ong, Yu Hui.
format Final Year Project
author Ong, Yu Hui.
author_sort Ong, Yu Hui.
title Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective.
title_short Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective.
title_full Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective.
title_fullStr Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective.
title_full_unstemmed Value-at-risk models under basel accord & Europe debt crisis : a Singapore perspective.
title_sort value-at-risk models under basel accord & europe debt crisis : a singapore perspective.
publishDate 2012
url http://hdl.handle.net/10356/49377
_version_ 1681035060633927680