摘要
准确估计组合内资产收益相关性是构建投资组合、定价衍生品以及风险管理的关键。引入波动择时策略从组合动态调整效率角度比较两类组合相关性估计模型的应用价值,并且利用不重叠的Block Bootstrap抽样对原始数据进行模拟,以期获得更可信的实证结论。结果表明根据基于高频数据的"已实现"模型对组合进行动态调整较静态组合会获得更多收益,而根据DCC-GARCH模型调整组合反而会有损失,说明"已实现"模型比DCC-GARCH模型更具有应用价值。
An accurate estimation of correlation of assets in the investment portfolio is the key of building portfolio, pricing derivatives and risk management. Volatility timing is introduced and the practical application of the two correlation estimation methods is compared from the perspective of dynamical portfolio allocation efficiency. The original data is simulated by the Block Bootstrap method in order to get more creditable result. The result indicated that, if we adjust the portfolio according to the RV model based on the high-frequency data,we will get more return, than static portfolio. However, if we adjust the portfolio according to the DCC-GARCH model we will loss. The RV model has more practical application than the DCC-GARCH model.
出处
《中国管理科学》
CSSCI
北大核心
2009年第1期1-6,共6页
Chinese Journal of Management Science
基金
国家自然科学基金资助项目(70771076)
国家杰出青年科学基金资助项目(70225002)
