Table 3: Estimation of the MGARCH model.

Full BEKK(1,1,1) Diagonal BEKK(1,1,1)CCC-GARCH(1,1) DCC-GARCH(1,1)

0.095 0.084 0.007 0.005
(0.008)(0.023)(0.003)(0.023)
0.089 0.068 0.012 0.008
(0.014)(0.019)(0.002)(0.001)
0.099 0.072 0.012 0.007
(0.015)(0.019)(0.003)(0.002)
0.062 0.046 0.066 0.064
(0.010)(0.014)(0.009)(0.008)
0.064 0.041 0.075 0.040
(0.011)(0.012)(0.010)(0.004)
0.0220 −0.007 0.078 0.064
(0.001)(0.002)(0.009)(0.008)
0.103 0.214 0.932 0.934
(0.022)(0.030)(0.009)(0.006)
0.040 0.198 0.901 0.948
(0.096)(0.032)(0.009)(0.007)
−0.071 0.191 0.904 0.916
(0.118)(0.028)(0.013)(0.010)
0.048 0.969 0.666 0.042
(0.023)(0.009)(0.011)(0.005)
0.151 0.972 0.666 0.942
(0.105)(0.010)(0.019)(0.008)
−0.113 0.973 0.429
(0.137)(0.008)(0.016)
0.037
(0.023)
−0.032
(0.123)
0.131
(0.125)
0.897
(0.013)
−0.019
(0.069)
0.007
(0.081) —     —       —    
0.001
(0.013)
0.849
(0.072)
−0.021
(0.082)
0.011
(0.010)
0.072
(0.068)
0.870
(0.083)

−756.715−743.369−758.793−760.490
−621.363−511.044−699.117−698.454
−707.114−768.169−781.993−787.756
646.262487.183141.769151.019
[0.000][0.000][0.481][0.399]
423.541361.06085.24299.283
[0.000][0.000][0.948][0.797]

Notes: The numbers in parenthese is standard error, is the correlation coefficient, and , and are, respectively, the Akaike, Bayesian and Hannan-Quinn information criterion. and are, respectively, the 12th order multivariate Portmanteau tests for serial correlation in the standardized and squared standardized residuals (null Hypothesis: no serial correlation). The numbers in brackets are the P-values.