Table 3.
USA |
China |
Hong Kong |
Japan |
South Korea |
|
---|---|---|---|---|---|
SP500 | CSI300 | HS | N225 | KS | |
0.0393** | 0.0398** | 0.02778 | 0.0403 | 0.0183 | |
(0.0170) | (0.0192) | (0.0195) | (0.0306) | (0.0287) | |
−0.0799*** | −0.0154 | −0.0034 | 0.0195 | −0.0194 | |
(0.0307) | (0.0292) | (0.1577) | (0.0241) | (0.0291) | |
−0.0514*** | 0.1874*** | 0.0982*** | 0.0157 | −0.1876*** | |
(0.0081) | (0.0627) | (0.0117) | (0.0359) | (0.0353) | |
−0.3234*** | −0.0856*** | −0.1327*** | −0.2078*** | −0.1491*** | |
(0.0412) | (0.0193) | (0.0199) | (0.0195) | (0.0289) | |
0.8255*** | 0.7358*** | 0.8121*** | 0.7354*** | 0.5413*** | |
(0.0204) | (0.0297) | (0.0392) | (0.0307) | (0.0874) | |
0.0931*** | 0.1529*** | 0.0594 | 0.0257 | 0.2205*** | |
(0.0118) | (0.0291) | (0.0399) | (0.0355) | (0.0421) | |
P1 | 0.2458*** | −0.2907*** | −0.1859*** | 0.1521*** | 0.2294*** |
(0.0526) | (0.0617) | (0.0499) | (0.0397) | (1.7665) | |
P2 | 0.1419*** | −0.2069*** | 0.0305 | −0.0343* | 1.7916*** |
(0.0271) | (0.0322) | (0.0181) | (0.0216) | (0.4516) | |
P3 | 0.1549*** | – | −0.2055*** | 0.2745*** | 0.4522*** |
(0.0165) | (0.0409) | (0.0914) | (0.1093) | ||
P4 | – | – | 0.1374*** | 0.1594*** | – |
(0.0215) | (0.0514) | ||||
P5 | – | – | – | −0.0878*** | – |
(0.0246) | |||||
Q2(20) | 11.631 | 9.9585 | 13.372 | 14.131 | 18.852 |
[0.967] | [0.8734] | [0.9012] | [0.6985] | [0.5172] | |
ARCH (20) | 0.4914 | 0.4125 | 0.7025 | 0.8257 | 0.9292 |
[0.8791] | [0.9547] | [0.8269] | [0.6837] | [0.5492] | |
AIC | 2.3944 | 3.3976 | 3.1082 | 2.9590 | 2.5809 |
BIC | 2.3982 | 3.3877 | 3.1458 | 3.0004 | 2.6147 |
LL | −1588.673 | −2412.363 | −2124.827 | −2037.975 | −17.588 |
BDS | −1.1125 | −0.1878 | 0.8245 | 0.5908 | −0.8531 |
[0.3175] | [0.8277] | [0.4575] | [0.5107] | [0.3295] |
Notes: Lagrange Multiplier test for heteroskedasticity is the ARCH (20). Ljung-Box Q-statistics of order 20 is the Q2 (20). The fitting of models is tested by the Akaike information criterion, Bayesian information criterion, and Log-likelihood noted AIC; BIC; and LL, respectively. *** **and *denote significance at 1%, 5% and 10%, respectively. The hypothesis that series are i.d.d is tested through the BDS test. Round brackets are used for the standard errors and square brackets for p values.