基于Kalman滤波算法的基金动态资产配置能力评价方法

构建了一个可以用于分析开放式基金动态资产配置结构的模型, 并利用Kalman滤波算法对我国54支开放式基金近5年的时间序列样本进行了实证检验. 结果显示: 该模型具有时变特征的状态参数设定能够较好地跟踪和近似拟合现实中基金对股票和债券资产配置结构的变动. 进一步扩展了模型, 提出了直接评价基金在股票和债券资产之间进行动态配置能力的方法. 实证结果显示: 该方法在分析基金投资业绩表现的效果上要优于传统的Treynor-Mazuy模型, 而且能够识别出我国开放式基金总体上具有资产配置能力, 基金长期投资业绩主要来源于资产配置决策.

Evaluating dynamic asset allocation ability of funds within Kalman filter framework

This paper constructs a model for measuring the asset allocation structure of China’s open-end funds.Based on this model,an empirical analysis within Kalman filter framework is conducted on the investment performance of 54 open-end funds within the recent 5 years.Results show the model that is characterized by time varying state parameters can well capture the real asset allocation strategies of these funds.We further extend our model to measure the dynamic asset allocation ability of the open-end funds. Empirical results show that our method outperforms the traditional Treynor-Mazuy model in evaluating the investment performance of funds open-end funds,and that open-end funds on the whole shows the strong ability in asset allocation,and that their long-run investment performance can mainly be attributed to their efforts on asset allocation.