中国证券市场中动态资产配置绩效的实证分析

An Empirical Analysis of Dynamic Asset Allocation Performance in Chinese Stock Market

采用"上证180"指数的周收益率数据,对基于随机规划方法的多阶段动态资产配置模型的投资绩效进行实证分析.研究结果表明:情景元素生成模型对动态资产配置模型的投资绩效起着关键性的作用,决策周期和资产配置频率等其他因素,只有在选择了适当的情景元素生成模型后,才能对投资绩效产生有利的影响,而中国证券市场目前的市场特征决定了采用反转策略的情景元素生成模型能产生较好的投资绩效;其次,动态资产配置模型一般在中长期投资中才能产生较好的投资收益;同时,最优的资产配置周期依赖于决策周期的时间跨度.

A test was conducted on the basis of weekly return rate of SSE180 index publlsned by Shanghai Security Exchanges to analyze investment performance of dynamic asset allocation model based on stochastic programming. The test results indicate that scenario generation model plays a crucial role in the dynamic asset allocation model and is the key factor affecting the investment performance of dynamic asset allocation. Other parameters, such as decision period or asset allocation frequency do not work without a proper scenario generation model. The scenario generation model based on ＂reverse strategy＂ leads to high performance given current Chinese stock market＇s features. Secondly, the dynamic asset allocation only achieves higher performance in the long term investment. Finally, the optimal number of asset allocation stages depends on the time span of the investment and decision periods.