Font Size: a A A

The Application Of Quantile Regression In The Financial Risk Management

Posted on:2012-12-29Degree:MasterType:Thesis
Country:ChinaCandidate:M Y ZhangFull Text:PDF
GTID:2219330374453543Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
At present, most of the world's people use the risk management of the standard is the 1990s flourishing after the VaR that this new type of risk management tool method is to use said easy, a specific numerals future value at risk, thus easy management and analysis data, and easy to understand, but the actual financial market is very complex, and in recent years international financial crisis of frequent, make people metric volatility increasing financial market risks, and demand a more accurate risk management tools. Therefore, appeared various improvement method of calculating VaR. Such as history simulation method and monte carlo simulation method, t GARCH model and so on, however-the above methods there are insufficient:history simulation method requirement and the actual simulated the actual situation of financial market changes inconsistencies and sometimes even sent out a lot of, monte-carlo simulation calculation complexity, and large amount of calculation, need to calculate the time is long, if data generated for pseudo random sequence, it may appear conclusion GARCH model error, t-need to establish in on the financial revenue data specific distribution basis. So we consider to need to use a new way to calculate VaR values. By mentioned VaR definition, can will VaR as financial income distribution of a certain points of digits. Therefore consider using points regression method to solve VaR median value. Use var to the model~1996 in shanghai and hong kong stock market was evolving patterns of risk and empirical research granger causality tests found by shanghai and hong kong stock market risk of the changes in the interactive relationship. the model parameters is no use reasoning and mcmc bayes the simplex method and used with quasi-newton method of searching, and mixed with matlab software fminsearch, fminnunc the function. And indirect GARCH(1,1) model for comparison. so the model for the shenzhen market is a rather ideal model of can better express the market mechanism of digits changed, the market risk model. in practice, the evolution of the shenzhen market investors have already exposed.
Keywords/Search Tags:Value at Risk, financial risks, quantile regression, the multiplication
PDF Full Text Request
Related items