Can Chinese security companies use Value at Risk (VaR) to measure market risk they faced: an empirical study?

Tan, Xiao (2006) Can Chinese security companies use Value at Risk (VaR) to measure market risk they faced: an empirical study? [Dissertation (University of Nottingham only)] (Unpublished)

[thumbnail of 06MAlixxt3.pdf] PDF - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (958kB)

Abstract

Many security companies have been launched since the establishment of the Chinese stock market, and the market has been steadily growing up. Until recently, Chinese stock market collapse despite the countrys economy growth, which cause many security companies and fund companies to go bankruptcy. Many factors have been believed directly and indirectly to incur this problem, but the most important reason is the poor management of market risks. Value at risk (VaR) models have become a very popular tool for measuring the market risk of a portfolio of financial assets, so the main objective of this paper is to analyze the market risk faced by the security companies, and then find an appropriate model of VaR to measure the market risk for Chinese securities companies. Based on the modern market risk management theory, this paper analyzes the current market risks Chinese securities company is facing and presents the Chinese securities company risk management system. Next, it introduces VaR method and conducts an empirical research of VaR application based on the Chinese financial industry current situation. The main finding of this research suggests that: (1) Chinese security companies need to enhance the market risk management system, and realize the importance of establishing VaR concept to measure the market risk. (2) The rate of return of Chinese stock market does not obey the normal distribution, and it has the peak and fat tail characteristic and the high stage ARCH effect. (3) The GARCH model can be used to portray the above characteristics well. The Kupiec test result showed that under 95% confidence level, the GARCH model passed the back-testing. Therefore, empirical research has proved that GARCH model can be incorporated into VaR to measure market risk by Chinese security companies.

Item Type: Dissertation (University of Nottingham only)
Keywords: VaR, Market risk, GARCH model
Depositing User: EP, Services
Date Deposited: 17 Jan 2007
Last Modified: 30 Dec 2017 15:04
URI: https://eprints.nottingham.ac.uk/id/eprint/20245

Actions (Archive Staff Only)

Edit View Edit View