金融数据处理(波动溢出).docVIP

金融数据处理(波动溢出) 由于Eviews在对波动溢出效应估计方面，只能得到对角矩阵。因此在用BEKK模型时，将使用S-Plus配合来估计波动溢出效应。 实验步骤： 第一步 检查平稳性 第二步 协整检验 第三步 Granger因果检验 第四步 VAR 第五步 VEC 第六步 BEKK  第一步 检查平稳性 Null Hypothesis: RHIS has a unit root Exogenous: Constant, Linear Trend Lag Length: 0 (Automatic based on SIC, MAXLAG=21) t-Statistic ??Prob.* Augmented Dickey-Fuller test statistic -35.20731 ?0.0000 Test critical values: 1% level -3.966233 5% level -3.413815 10% level -3.128983 *MacKinnon (1996) one-sided p-values. Augmented Dickey-Fuller Test Equation Dependent Variable: D(RHIS) Method: Least Squares Date: 11/20/11 Time: 16:28 Sample (adjusted): 1/06/2006 11/08/2010 Included observations: 1124 after adjustments Variable Coefficient Std. Error t-Statistic Prob.?? RHIS(-1) -1.050214 0.029829 -35.20731 0.0000 C 0.000497 0.001236 0.402269 0.6876 @TREND(1/04/2006) -6.72E-08 1.90E-06 -0.035341 0.9718 R-squared 0.525112 ????Mean dependent var -1.09E-06 Adjusted R-squared 0.524264 ????S.D. dependent var 0.029972 S.E. of regression 0.020673 ????Akaike info criterion -4.917329 Sum squared resid 0.479076 ????Schwarz criterion -4.903918 Log likelihood 2766.539 ????Hannan-Quinn criter. -4.912261 F-statistic 619.7774 ????Durbin-Watson stat 1.999385 Prob(F-statistic) 0.000000  可知，rhis系列没有单位根，即是平稳的。  Null Hypothesis: RNQI has a unit root Exogenous: Constant, Linear Trend Lag Length: 0 (Automatic based on SIC, MAXLAG=21) t-Statistic ??Prob.* Augmented Dickey-Fuller test statistic -36.79906 ?0.0000 Test critical values: 1% level -3.966233 5% level -3.413815 10% level -3.128983 *MacKinnon (1996) one-sided p-values. Augmented Dickey-Fuller Test Equation Dependent Variable: D(RNQI) Method: Least Squares Date: 11/20/11 Time: 16:27 Sample (adjusted): 1/06/2006 11/08/2010 Included observations: 1124 after adjustments Variable Coefficient Std. Error t-Statistic Prob.?? RNQI(-1) -1.094133 0.029733 -36.79906 0.0000 C -0.000443 0.001071 -0.413305 0.6795 @TREND(1/04/2006) 1.00E-06 1.65E-06 0.609665 0.5422 R-squared 0.547102 ????Me