【摘要】 引入灰色模型和符号时间序列分析方法,与马尔科夫模型方法相结合,提出了一种新的预测金融波动的方法。首先将波动序列符号化,然后建立灰色马尔科夫模型,不仅能减小影响预测精度的误差,而且能利用马尔科夫模型来调整误差,使结果更加精准。采用上海证券交易所综合指数2007—2010年间隔为5分钟的高频数据为样本,对已实现波动序列进行实证分析,成功预测了下一时点波动值所处的区间,并验证了该方法的可行性和有效性。更多还原

【Abstract】 Through the introduction of Gray model symbolic time series analysis,a new method of forecasting financial volatility was put forward combined with the Markov model method. Firstly,the volatility series need to be symbolized. Secondly, Gray- Markov model was built. Not only can the errors impacting prediction accuracy be reduced,but also the Markov model make the results more accurate by adjusting them. With high frequency data whose sampling intervals are 5 minutes from Shanghai Stock Exchange Composite Index from 2007 to 2010 as a sample,analyzing the realized volatility sequence empirically,the range of the fluctuations of the next time are predicted successfully,which verified the feasibility and effectiveness of the method.更多还原