A Study On Financial Market Risk Measurement On Condition Of Heavy-tailed Distributions

After 1970s, the collapse of fixed price system, financial innovation and the globalization of financial market increasingly aggravate the volatility of financial market. Market risk has been the most important form of financial risks. The ability of market risk management has been one of the essential capacities gaining the competitive privilege for financial institutions.Many empirical researches have indicated that distributions of return of risky securities are heavy-tailed on condition of which the probability of extreme cases is much bigger than on condition of normal distribution. As a result, on condition of heavy-tailed distributions, it is more important to manage financial market risk.Meanwhile, without considering the characters of heavy-tailed distributions, risk management models will certainly underestimate risk and cause seriously effects. However, on condition of heavy-tailed distributions, VaR dissatisfies convexity and subadditivity, and can't measure market risk validly.Consequently, studying financial market risk management models on condition of heavy-tailed distributions has not only theoretic value but also practical value, and is the hotspot in researches of financial theory.Risk measurement is the foundation and hardcore of financial market risk management. This thesis will study the financial market risk measurement on condition of heavy-tailed distributions. There are five chapters in this thesis.Chapter 1 introduces the basic definitions and theories of financial risk, financial risk management and financial market risk management.Chapter 2 studies the financial market risk measurements on condition of heavy-tailed distributions. Proposes to define and measure risk in terms of distribution of the outcomes of uncertain events as well as references; introduces the coherent risk measurement; introduces ES, TCE and CVaR, redefines them in a uniform and normative terms and demonstrates the differences and relations of them.Chapter 3 introduces the methods of computing rate of return in finance, tests the heavy-tailed character of rate of return of stocks in China using four test methods. The results indicate that distributions of rate of return in China is surely heavy-tailed.Chapter 4 uses empirical analysis to test the convexity and subadditivity of VaR, ES and TCE, and test the validity of VaR and ES by back test in China stock market.Chapter 5 summarizes the whole article and puts forward some viewpoints for further researches based on the conclusion of this thesis.