回归的C(1)和C（2）的数值很奇怪，为什么是一样的呢？是不是方法不对呢？
一、原始的VAR模型：
XINMAOYI = 1.27281815244*XINMAOYI(-1) + 2.27938808567*WANGANG(-1) - 61401.8641332
WANGANG = 0.0616745685426*XINMAOYI(-1) + 0.0258674716182*WANGANG(-1) + 5815.11662756


二、对应的SVAR模型：

Structural VAR Estimates   
Date: 04/27/09   Time: 06:29   
Sample (adjusted): 1991 2007   
Included observations: 13 after adjustments   
Estimation method: method of scoring (analytic derivatives)   
Failure to improve after 1 iterations   
Structural VAR is over-identified (1 degrees of freedom)   
   
Model: Ae = Bu where E[uu']=I   
Restriction Type: short-run pattern matrix   
A =     
1 C(2)   
C(1) 1   
B =     
1 0   
0 1   
WARNING: B matrix is fixed (structural innovation variances not estimated)!!!   
   
Coefficient Std. Error z-Statistic Prob.  
   
C(1)  0.100000  1.379663  0.072481  0.9422
C(2)  0.100000  1.379663  0.072481  0.9422
   
Log likelihood  -5.71E+09   
LR test for over-identification:     
Chi-square(1)   1.14E+10  Probability  0.0000
   
Estimated A matrix:   
1.000000  0.100000   
0.100000  1.000000   
Estimated B matrix:   
1.000000  0.000000   
0.000000  1.000000