Dependent Variable: LNDL
Method: Least Squares
Date: 08/03/10 Time: 12:19
Sample: 2005Q1 2010Q2
Included observations: 22
Variable Coefficient Std. Error t-Statistic Prob.
C 0.734136 0.206469 3.555668 0.0020
LNCZ 0.593968 0.026176 22.69113 0.0000
R-squared 0.962609 Mean dependent var 5.413502
Adjusted R-squared 0.960739 S.D. dependent var 0.239944
S.E. of regression 0.047543 Akaike info criterion -3.167849
Sum squared resid 0.045207 Schwarz criterion -3.068663
Log likelihood 36.84634 F-statistic 514.8874
Durbin-Watson stat 1.250978 Prob(F-statistic) 0.000000
d-w 临界值1.24,进行BG检验(滞后2阶),结果如下:
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 1.371333 Probability 0.279043
Obs*R-squared 2.908915 Probability 0.233527
Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 08/03/10 Time: 12:33
Presample missing value lagged residuals set to zero.
Variable Coefficient Std. Error t-Statistic Prob.
C 0.004396 0.220682 0.019921 0.9843
LNCZ -0.000575 0.028094 -0.020458 0.9839
RESID(-1) 0.376880 0.255002 1.477953 0.1567
RESID(-2) -0.253143 0.284702 -0.889150 0.3857
R-squared 0.132223 Mean dependent var -9.68E-16
Adjusted R-squared -0.012406 S.D. dependent var 0.046397
S.E. of regression 0.046684 Akaike info criterion -3.127852
Sum squared resid 0.039230 Schwarz criterion -2.929480
Log likelihood 38.40637 F-statistic 0.914222
Durbin-Watson stat 1.792460 Prob(F-statistic) 0.453776
Q-stats 结果
Date: 08/03/10 Time: 12:38
Sample: 2005Q1 2010Q2
Included observations: 22
Autocorrelation Partial Correlation AC PAC Q-Stat Prob
. |* . | . |* . | 1 0.141 0.141 0.5025 0.478
. | . | . | . | 2 -0.031 -0.052 0.5281 0.768
. |* . | . |* . | 3 0.169 0.186 1.3252 0.723
. |* . | . | . | 4 0.067 0.012 1.4577 0.834
. *| . | . *| . | 5 -0.098 -0.096 1.7551 0.882
. | . | . | . | 6 -0.013 -0.009 1.7604 0.940
. | . | . | . | 7 0.031 0.009 1.7934 0.970
. | . | . | . | 8 0.004 0.031 1.7940 0.987
. | . | . | . | 9 0.011 0.022 1.7988 0.994
. | . | . | . | 10 -0.010 -0.031 1.8033 0.998
. | . | . | . | 11 0.000 -0.002 1.8033 0.999
. | . | . | . | 12 0.000 -0.006 1.8033 1.000
该检验结果能否证明残差具有自相关性质?或者可以使用其他什么方法来判断自相关?