Testing various methodology of VaR under crisis of 2008

by

Acknowledgement

I would take this opportunity to thank my research supervisor, family and friends for their support and guidance without which this research would not have been possible.

DECLARATION

I, [type your full first names and surname here], declare that the contents of this dissertation/thesis represent my own unaided work, and that the dissertation/thesis has not previously been submitted for academic examination towards any qualification. Furthermore, it represents my own opinions and not necessarily those of the University.

Signed __________________ Date _________________

Abstract

Over the past decades portfolio and risk management techniques had adapted to increasingly complex financial instrument. Within the different forms of financial risk measurement tools, Value at Risk (VAR) which provides the most expediency measurement from the adverse market movements, is now widely accepted as a fundamental tool for risk management and it has become a standard benchmark for measuring financial risk since the 1990s. This dissertation primarily focuses on using the newly created Australian implied volatility as an input for Value-at-Risk models during the financial crisis period and then compares the testing results with other two different volatility inputs based on two back-testing methods. The results show that during the financial crisis period straight forward volatility forecasts based on Australian implied volatility do not provide meaningful volatility information in VAR models, and this was however fine in most cases when using RiskMetrics and GJR-GARCH as volatility forecasts methods. This indicates that the models? performances can be deteriorating in challenging trading environments, and in order to get protection against credit risk, operational risk and liquidity risk, the risk managers or investors should appropriate use of VAR

Table of Contents

TABLE OF CONTENTS5

CHAPTER 1: INTRODUCTION6

Introduction6

Background of the Study7

Research Aims and Objectives9

CHAPTER 2: LITERATURE REVIEW10

Risk Management12

The importance of analyze risks and measure risks12

Financial Risk Measurement13

Value at Risk14

Different methods of computing VAR15

Correlation Method16

Historical Simulation Method16

Monte Carlo simulation method17

CHAPTER 3: DATA AND METHODLOGY18

Research Methodology24

Volatility Analysis25

Efficiency and unbiasedness of implied volatility26

Lagged implied volatility26

CHAPTER 4: EMPERICAL ANALYIS29

Implied volatility and Realized Volatility Regression Analysis29

Volatility forecasts and back testing on VAR30

In Sample Test32

Back-testing on In-Sample Results34

CHAPTER 5: CONCLUSIONS AND RECOMMENDATION37

REFERENCES39

CHAPTER 1: INTRODUCTION

Introduction

The current worldwide financial crisis has revealed major deficiencies in the existent financial risk measures. For example, based on one-day-ahead value-at-risk (VaR) forecasts, which constitutes the focus of interest of the present paper, JPMorgan Chase reported 5, Credit Suisse 7 and UBS 16 exceedances in the 3rd quarter of 2007, which required a maximum of 0.63 exceedances for a probability level of 1% (Jorion, forthcoming). Considerable progress has been made over the last decade to quantify financial risks by means of elaborate econometric tools. However, experience with the performance of these methodologies in rough times was missing until the subprime crisis started to shake the world financial markets (Jorion, 1995, pp. 507). Most of the risk models are based on a set of assumptions which may be more or less carefully validated most of the time, but may contribute to an increase in systemic risk if the model assumptions fail to be true for other ...