中文摘要：

研究多元市场间的相关性对构建市场投资组合进而有效规避风险具有重要现实意义。将股票、基金、国债、期货、货币、外汇和现货市场纳入统一框架，以2010年10月1日至2014年3月31日的HS300指数、上证基金指数、国债指数、燃油期货指数、Shibor隔夜拆借利率、欧元兑人民币中间价和原油商品指数为样本，在结合GJR和EVT对各自边缘分布进行估计的基础上，运用R-Vinecopula、D-Vinecopula和C-Vinecopula3种模型综合探讨中国不同金融市场之间的净相关关系，并详细分析所有两两市场间的非条件相关性及其在一个市场条件下的条件相关性。实证结果表明，中国金融市场间表现出厚尾相关性和非对称相关性特征，3种不同Vinecopula模型对中国金融市场的拟合效果没有显著差异；多数两两市场间具有较高的非条件相关性，通过逐一分析除两两市场外的第3个市场时其相关性的影响得知，二者在某些市场条件下的条件相关性仅为其非条件相关性的20％以下，因而选择对应市场构建三元投资组合可以避免仅在两个市场同向投资时其市场价格同时下跌的风险；熊市时要避免直接在非条件相关性及条件相关性较强的市场间同向投资，应通过选择非条件相关性较低的两个市场、条件相关性表现独立或较低的多个市场构建投资组合，从而规避不同市场价格同时下跌的风险。

英文摘要：

There is significance for avoiding investment risk through choosing market investment portfolio with studying the corre- lation among multi-markets. Taking the stock, fund, bond, future, money, exchange and spot markets in China into a frame- work, with the indices of HS300, Shanghai Stock Exchange Fund, Bond Index, Fuel Oil Future, Shibor overnight interest rate, central parity rate of Euro against Yuan and crude oil commodity between October 1,2010 and March 31,2014 as samples, this study first modeled its marginal distribution with GJR-EVT method, and then analyzed the net correlation among different finan-cial markets with the R-Vine copula, D-Vine copula and C-Vine copula methods, last analyzed the all non-conditional correlation between any two markets and its conditional correlation on the other one conditional market. The main empirical results show that there is significant asymmetric and fat tail correlation characteristics between the different financial markets in China, and the three Vine copula models are not significant different in modeling the Chinese financial markets; There exists high non-condition- al correlation between most two of the markets, but their conditional correlation is lower than the 20 percent of non-conditional correlation after taking every other market into account. Composing the triple-portfolio with the corresponding market can avoid the market prices falling at the same time when investment are in the binary-markets; In bear market period, avoiding only long or short investment in the markets with high unconditional or conditional correlation, it could hedge the risk of different market prices fall at the same time through selecting these markets with low non-conditional, or the markets with low or independent con- ditional correlation to build investment portfolio.