Font Size: a A A

An Empirical Analysis Of Financial Crisis Contagion Based On Change Point Testing Method Of Copula

Posted on:2015-08-28Degree:MasterType:Thesis
Country:ChinaCandidate:D MaFull Text:PDF
GTID:2309330461960619Subject:Applied statistics
Abstract/Summary:PDF Full Text Request
Financial contagion is an important issue explored by domestic and foreign researchers.To further study the contagion effect of the financial crisis, we must be familiar with the internal correlations of the financial markets, copula function is an important tool for studying the correlation between financial variables.In this paper, copula theory and its applications in Financial contagion are studied intensively. As a result of those researches, several multivariate financial time series models based on copula theory are proposed. Methods of constructing models in stages and applications of copula models in Financial contagion are also investigated systematically.The major work of this paper could be summarized as follows:Firstly, stock returns are decomposed into new sequences using Kernel density estimation, and tested if they were suitable for the Marginal distributions of copula models with Kolmogorov-Smiynov test and Independence test. Then, the best copula models are measured by the squared Euclidean distance method and the Klugman Parsa method. Finally, taking Marginal distributions as study object of Financial contagion effect, variable structure copulas are established, and estimating the parameters and the tail correlation coefficients of each function.Based on the methods above, using the daily closing price of the S&P500, Russia’s RTS index,Korea KOPSI index, Chinese Hong Kong’s Hang Seng Index, CSI 300 Index, Mumbai SWNSEX300 index, the Nikkei 225 index (NIKKI),the Greek ASE index, Britain’s Financial Times Index (FTSE) from January 3,2006 to January 10,2014 as the sample data, Financial contagion effect between S&P 500 and each one of the rest stock time series is deeply analyzed in this paper. Results show that:(1)S&P500 is consistent with the trend of other indices other than the Hang Seng Index, the tail-dependence coefficient of each copula model shows that the stock markets of Germany,Russia,South Korea,Greece, the UK and Greece have strong synergistic effect with U.S. stock market, financil crisis can be easily transmitted between those markets.(2)Change points exist in every copula models between S&P 500 and other stock index, and the time of change points are almost unanimously with time of the global financial crisis.According to the changing points of the copulas, stock indexes in this paper are all affected by the United States subprime financial crisis more or less, however, the European debt crisis has a limited impact.
Keywords/Search Tags:variable structure Copula, financial contagion, kernel density estimation
PDF Full Text Request
Related items