Font Size: a A A

Research On Risk Measurement Of Financialassets Based On Copula

Posted on:2014-05-31Degree:MasterType:Thesis
Country:ChinaCandidate:J H WenFull Text:PDF
GTID:2269330425961012Subject:Probability theory and mathematical statistics
Abstract/Summary:PDF Full Text Request
The applied research, which is based on financial assets portfolio riskmeasurement of Copula, is an advanced topic in the area of financial riskmanagement.With the acceleration of economic globalization, the domestic financialmarket has become increasingly complex and diverse. The correlative patterns ofmarket risk measurement show features such as nonnormality, asymmetry and taildependence. Compared with previous analysis methods based on normal distributionassumption, copula function has its unique advantages on the correlative analysis offinancial risk.Combining copula theory with the stochastic volatility model ofdescribing marginal distribution, the paper builds t-Copula-SV-t model to makecorrelative analysis of the two main Chinese energy stocks, and observes the risk ofstock using risk measurement theory to calculate the value at risk.In theory, the paper firstly introduces the research status of the stochasticvolatility model and copula theory in the overseas and domestic research. And itbriefly presents the application of Bayesian measurement tool. It makes detailedelaboration of copula function and stochastic volatility model by chapters, includingdefinition, classification, parameters estimation and inspection of the model.Then theconcept of risk value and its calculated methods under different conditions arepresented to prepare for latter risk measurement theoretically.In empirical aspect, Shanghai and Shenzhen energy stocks are selected asresearch samples. The first thing we do is to analyze the basic statisticalcharacteristics of data.Stochastic volatility model can be used to build marginaldistribution of variables. By comparing and testing, we can select the most suitableSV-t model as marginal distribution to build different kinds of copula models.It seemsthat t-Copula model’s imitative effect is better than other models from the densityfunction figures and theoretical evaluation. On the basis of the model, we cancalculate the value at risk of assets portfolio.Finally, the paper makes a conclusion of the application of copula theory inChinese financial market correlation research and risk measurement analysis, and puts forward to our further research problems and prospects.
Keywords/Search Tags:risk measurement analysis, the stochastic volatility model, Copulafunction, MCMC simulation, Bayesian
PDF Full Text Request
Related items