兰州大学机构库 >数学与统计学院
我国股市与国债市场的相关性研究——基于极值理论与Copula函数
Alternative TitleThe Study on the Relativity Between the Stock Market and Government Bond Market in China——Based on Extreme Value Theory and Copula Function
韩玉婷
Thesis Advisor焦桂梅
2018-04-04
Degree Grantor兰州大学
Place of Conferral兰州
Degree Name硕士
Keyword股票市场 国债市场 ARMA-GARCH模型 极值理论 Copula函数
Abstract

随着金融市场日趋多元化、复杂化,金融市场之间的相关性也随之增强。对投资者来说,股票市场和国债市场作为两个非常重要的金融市场组成部分,如何配置和调整其中的资金比例就显得尤为重要,并且研究二者的相关性也直接影响相关经济政策的决策,可以有效满足投资主体对于风险的管理,优化其资源配置,获得较高收益。

本文首先介绍了股市与国债市场,然后对股市与国债市场相关性的理论进行了阐述,最后介绍了相关模型的理论,包括模型、极值理论和理论,并对比了不同类型函数的特点。 实证部分,选取上证综合指数和上证国债指数的对数收益率,用来代表股票市场和国债市场的波动情况,并对它们经过一系列统计检验后,在残差分布为正态分布、分布和分布下分别建立模型,根据AIC信息准则,择优选择了模型对边缘分布进行拟合。接着引入极值理论,对模型的残差分布建立模型来优化边缘分布。最后,用正态函数、函数、函数、函数和函数对2个指数间的相关结构进行建模,根据检验法,得出函数为最优的函数,所以本文用函数描述上证综合指数和上证国债指数的相关关系。

Other Abstract

As financial markets become more diversified and complex, the correlation between financial markets increases. For investors, the stock market and government bond market as two very important part of financial market, how to configure and adjust the proportion of which is particularly important. Study the correlation of the two also directly affect the relevant economic policy decisions, which can effectively meet the main body of investment for risk management, optimize the allocation of resources, and get a higher income.

This article first introduces the stock market and government bond market correlation theory, and then the ARMA-GARCH model, the theory of extreme value and copulas connect systematically, and expounds several common Copula connect function in detail, points out that the different kinds of Copula connect function characteristics.

The empirical part selects the Shanghai composite index and the Shanghai Stock Exchange T-Bond Index to represent the stock market and the government bond market. After a series of statistical tests, ARMA-GARCH model is established under normal distribution、t distribution and GED distribution of residual error, and ARMA-GARCH(1,1)-GED is selected according to the AIC.Then the POT model is introduced to optimize the edge distribution.Finally, the correlation structure between two exponents is modeled by normal Copula、Student-t Copula、Gumbel Copula、Clayton Copula and Frank Copula.Through GOF test, the Gumbel Copula function is selected to describe the correlation between the two markets. 

URL查看原文
Language中文
Document Type学位论文
Identifierhttps://ir.lzu.edu.cn/handle/262010/224596
Collection数学与统计学院
Recommended Citation
GB/T 7714
韩玉婷. 我国股市与国债市场的相关性研究——基于极值理论与Copula函数[D]. 兰州. 兰州大学,2018.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Altmetrics Score
Google Scholar
Similar articles in Google Scholar
[韩玉婷]'s Articles
Baidu academic
Similar articles in Baidu academic
[韩玉婷]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[韩玉婷]'s Articles
Terms of Use
No data!
Social Bookmark/Share
No comment.
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.