Font Size: a A A

The Comparison Of Several Statistical Inference Methods Of VaR

Posted on:2016-08-09Degree:MasterType:Thesis
Country:ChinaCandidate:N WangFull Text:PDF
GTID:2309330464453764Subject:Applied Statistics
Abstract/Summary:PDF Full Text Request
The development of the world economy has entered the globalization and integration, the global financial market occupies an important position in the development of economic globalization, in the whole process of the development of the global financial market, one of the topic that attracts the most attention is the financial risk, in this context, to measure a country’s financial market development status of key point is its ability to control the financial market’s risk. VaR developed as the new financial risk management tools by global financial institutions, government and other social aspects in the recent 20 years, VaR is now is a major means of risk measurement in international financial markets, much of the financial investment in the field of risk control is one of the international standard, and the expected loss ES as a good complement of VaR, also attracted a lot of financial scholars research about it. Therefore, study of VaR and ES to measure in theory and practice of financial risk has important significance.Value at Risk and Expected shortfall, there are many kinds of financial risk measure value and the world expect losses used mainly by the historical simulation method, Monte Carlo method, the nonparametric estimation method and extreme value theory method, the calculation method of the four kinds of VaR and ES each have advantages, are calculated under a certain condition, for the development of the rapidly changing financial markets, which are volatile, must under the premise of effective reasonable conditions is calculated.This paper first briefly introduces the four kinds of VaR and ES calculation method and its main features, and then select two random white noise model to produce enough data, respectively with the four kinds of VaR and ES methods to calculate, the results compared with nominal level, analysis their respective comparison results, the real reason for existence results were discussed, at last, through to the empirical analysis of the actual data of the stock market, to explain the four kinds of VaR and ES the merits of the method to calculate the use effect.
Keywords/Search Tags:Value at Risk, Expected shortfall, Nonparametric estimation, Extreme value theory, Stochastic simulation
PDF Full Text Request
Related items