| It is well known, Yue’ bao received extensive attention of the social public, which is based on the third party payment platform supporting by Internet. And it had a profound impact for the classical business bank. As one of the financing market, the returns of Yue’bao has some volatility characteristics. In this paper, we discussed the volatility and the forecast of the returns of Yue’ bao. The 7 days of yield of Yue’ bao from 3 June of 2013 to 10 April of 2015 is studied for its time varying volatility. And ARIMA model was used to establish the mean equation, then the GARCH)1,1( model, the GJR-GARCH)1,1( model,the EGARCH)1,1( model, the APARCH)1,1( model was used for modeling the residual equation from the ARIMA model. The result indicates that the series has the nonstationary and the auto-correlation, and one order difference sequence has an obvious peak and fat tails, volatility clustering and conditional heteroskedasticity. Furthermore,the ARIMA-GARCH family models are established to the comparative study of model parameters and the prediction evaluation indexes based on those volatility characteristics.The analysis indicates that the hybrid model ARIMA-GJR-GARCH has the best prediction effect. Finally, the paper performs the cointegration test and the causality test and the graph of impulse response function with the 7 days of yield and the 10 thousand accrual, which indicates that there is a reliable cointegration relationship and causal relationship between them, and the 7 days of yield to the 10 thousand accrual have obvious positive impact. |
