基于VaR的我国开放式基金市场风险研究

发布时间：2018-04-25 03:09

本文选题：开放式基金 + VaR理论　；参考：《合肥工业大学》2014年硕士论文

【摘要】：近年来，伴随着经济全球化和金融一体化的进程，中国的金融市场取得了前所未有的发展，金融风险也呈现加剧现象。开放式基金作为一种专家管理的集合投资方式，已成为我国证券市场的主要参与者。如何度量其风险，已成为金融监管机构与基金持有者共同关心的问题。因此，研究开放式基金市场风险具有重要的理论和实践意义。本文针对后危机时期我国开放式基金市场，应用基于VaR理论的GARCH模型对我国开放式基金风险进行了实证研究。论文选取十五只基金的日收益净值数据，使用Eviews7.0计量分析软件对这十五只开放式基金时间收益序列进行综合分析，得出各自收益率分布情况，并利用ARCH-LM检验时间序列存在异方差、波动聚集性等特征。随后以单只基金为例选择合适的模型，再利用此模型在t分布和GED分布的假设下，建立了估计基金风险的VaR-GARCH模型，，对各只基金及不同类型基金的VaR值进行估计。最后在95%和99%置信水平下，运用Kupiec方法对VaR模型的准确性进行了返回检验。得出如下主要结论: t分布可以刻画基金收益率序列的尖峰厚尾特征，但是t分布的尾部较厚，根据GARCH(1,1)-t分布模型计算VaR值会高估真实的风险，根据GARCH(1,1)-GED分布模型计算的VaR值能较真实的反映基金风险；95%置信水平下，GARCH-t模型存在高估风险，而GARCH-GED模型存在低估风险；99%置信水平下，GARCH-t模型和GARCH-GED均存在高估风险。最后，基于实证研究的启示，结合我国开放式基金风险管理的现状，提出了相应的政策建议。
[Abstract]:In recent years, with the process of economic globalization and financial integration, China's financial market has achieved unprecedented development and financial risks are becoming more and more serious. As an expert management, the open fund has become the main participant in the securities market. How to measure its risk has become a financial supervision. Therefore, it is of great theoretical and practical significance to study the market risk of open-end funds.
In view of the open end fund market of China in the post crisis period, this paper uses the GARCH model based on VaR theory to carry out an empirical study on the risk of open-end fund in China. The paper selects the net daily income data of fifteen funds, and uses the Eviews7.0 econometric analysis software to comprehensively analyze the time income sequence of the fifteen open-end funds. The ARCH-LM test time series has the characteristics of heteroscedasticity and volatility aggregation. Then a single fund is used as an example to choose a suitable model. Under the assumption of t distribution and GED distribution, a VaR-GARCH model for estimating fund risk is established, and the VaR value of funds and different types of funds is also established. At the end of the 95% and 99% confidence levels, the accuracy of the VaR model is tested by the Kupiec method. The main conclusions are as follows: the t distribution can depict the peak and thick tail characteristics of the fund yield sequence, but the tail of the t distribution is thicker, and the GARCH (1,1) -t distribution model can overestimate the real risk, root, and the root of the GARCH (1,1) -t distribution model. According to the GARCH (1,1) -GED distribution model, the VaR value can reflect the risk of the fund more truly; under 95% confidence level, the GARCH-t model overestimates the risk, and the GARCH-GED model underestimates the risk; under the 99% confidence level, the GARCH-t model and GARCH-GED are all overestimated risk. Finally, the Revelation based on the empirical study, combined with our open-end fund The present situation of risk management puts forward corresponding policy recommendations.

【学位授予单位】：合肥工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51

【参考文献】