基于时变的我国开放式基金选股和择时能力定量分析

金融时间序列数据常常发现有波动聚集的现象即波动随时间变化的现象.选取2003年前发行的15只开放式基金的周数据作为样本,在T-M模型和H-M模型的基础上加入GARCH效应来分析基于时变的我国开放式基金的选股和择时能力,以求对这两种能力更精确的评估.实证结果表明,传统的T-M模型和H-M模型不仅高估了基金经理的选股能力,还高估了基金投资组合的系统风险.因此,为了去除这种高估的偏差,应该在评估基金绩效或基金经理选股、择时能力时在模型中考虑GARCH效应.

Quantitative analysis on selectivity and timing abilities of China open-end mutual fund under volatility clustering

Financial time series usually exhibit a characteristic known as volatility clustering. In this paper, we analyze open-end funds selectivity and market timing ability under volatility clustering , through adding GARCH(Generalized Auto-Regressive Conditional Heteroskedastic) model to T-M (Treynor and Mazuy) and H-M(Henriksson and Merton) models ,using the weekly returns of 15 open funds established before 2003. Conditioning on timing volatility, we find that traditional T-M and H-M models overestimate the fund managers' selectivity and the systemic risk of portfolio. Therefore, in order to measure more precisely funds performance or the selectivity and timing ability, GARCH effect should be considered.