基于TGARCH-M模型的股票型基金投资风格漂移动态识别及原因分析

Research on Dynamic Identification of Stock Funds Investment Style Drift and Causes Based on TGARCH-M Model

股票型基金投资风格随市漂移已成为一种常态，故动态识别基金风格漂移现象具有重要意义。本文构建了识别投资风格漂移的TGARCH—M模型。并选取2006年之前成立的开放式股票型基金作为样本，将2006～2015年的股市行情分为上涨、下跌、回调、震荡、再次上涨五个阶段，利用TGARCH—M模型对这五个阶段及整个期间的样本股票型基金投资风格漂移和收益率波动情况进行了动态验证。结果表明该模型能够较好识别投资风格漂移现象：占比85．71％的基金在长期均发生了风格漂移，在股市下跌阶段漂移现象更为严重，而在股市上涨阶段大部分基金可以坚持投资风格。同时风格漂移与收益率的波动大小无必然联系。最后，本文进一步分析了投资风格漂移的可能原因，并对监管投资风格漂移现象给出了相关建议。

As the investment style of stock funds often drifts, how to dynamically identify the style drift is of great significance to investors. This paper constructs TGARCH-M model to identify the fund investment style drift. It picks 14 open-end stock funds that established before 2006 as samples. We divide the stock market from 2006 to 2015 into five periods: up-trend period,down-trend period, consolidation period, shock period and the second up-trend period. We used TGARCH-M model to made dynamic verification on whether style drift occurred during the five periods and the whole section. We also investigates the return volatility. The results show that this model can preferably identify the investment style drift. About 85.7% of sample funds show the evidence of style drift during a long period, and the phenomenon of style drift is more common during the down-trend period. During the up-trend period, most samples could insist their style, and style drift did not necessarily means return volatility will become severe. Finally, this paper makes a further analyzes on why the investment style drift happened, and puts forward the policy recommendation about supervisor.