Estimating and forecasting value at risk and expected shortfall: a backstory of Malaysia stock market during global financial crisis / Ahmad Fauze Abdul Hamit

A myriad of factors leading to the calamity of the financial crisis have been widely examined. Yet, past studies merely focused on the major causes while ignoring risk mismanagement - another important aspect which could be detrimental in certain circumstances. This study driven by the latest intent...

Full description

Saved in:
Bibliographic Details
Main Author: Abdul Hamit, Ahmad Fauze
Format: Thesis
Language:English
Published: 2018
Subjects:
Online Access:https://ir.uitm.edu.my/id/eprint/37208/2/37208.pdf
https://ir.uitm.edu.my/id/eprint/37208/
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Universiti Teknologi Mara
Language: English
Description
Summary:A myriad of factors leading to the calamity of the financial crisis have been widely examined. Yet, past studies merely focused on the major causes while ignoring risk mismanagement - another important aspect which could be detrimental in certain circumstances. This study driven by the latest intention of Basel Committee to replace the application of VaR to ES due to the failure of VaR during the financial crisis. We examine the performance of VaR and ES by using various methods during calm and stormy period in Malaysia stock market. 24 daily return series from TOP30 FTSE Bursa Malaysia KLCI were collected spanning from January 2007 to June 2009 covering the stormy period of crisis, as well as from July 2010 to December 2016 covering the calm period. Our findings suggest that the application of VaR is still relevant and superior to ES in Malaysia stock market with the higher confidence level which is 99% is more preferable to be used. Additionally, we also found that VaR is very accurate during calm period by using HS and GARCH with normal distribution while FHS is more suitable on the calculation of ES. Furthermore, the chaotic market condition during financial crisis made the prediction become less accurate.