三变量误差修正模型：两个解释变量,用JJ协整检验后得到以下结果,如何看这个结果?怎样做误差修正模型?Date:01/13/12 Time:12:06\x05\x05\x05\x05Sample (adjusted):1989 2010\x05\x05\x05\x05Included observations:22 afte

三变量误差修正模型：两个解释变量,用JJ协整检验后得到以下结果,如何看这个结果?怎样做误差修正模型?Date:01/13/12 Time:12:06\x05\x05\x05\x05Sample (adjusted):1989 2010\x05\x05\x05\x05Included observations:22 afte
三变量误差修正模型：两个解释变量,用JJ协整检验后得到以下结果,如何看这个结果?怎样做误差修正模型?
Date:01/13/12 Time:12:06\x05\x05\x05\x05
Sample (adjusted):1989 2010\x05\x05\x05\x05
Included observations:22 after adjustments\x05\x05\x05\x05
Trend assumption:Linear deterministic trend\x05\x05\x05\x05
Series:LNTFP LNOFDI LNS \x05\x05\x05\x05
Lags interval (in first differences):1 to 1\x05\x05\x05\x05
\x05\x05\x05\x05
Unrestricted Cointegration Rank Test (Trace)\x05\x05\x05\x05
\x05\x05\x05\x05
Hypothesized\x05\x05Trace\x050.05\x05
No.of CE(s)\x05Eigenvalue\x05Statistic\x05Critical Value\x05Prob.**
\x05\x05\x05\x05
None *\x05 0.664885\x05 41.44770\x05 29.79707\x05 0.0015
At most 1 *0.470765\x05 17.39553\x05 15.49471\x05 0.0256
At most 2\x05 0.143056\x05 3.396423\x05 3.841466\x05 0.0653
\x05\x05\x05\x05
Trace test indicates 2 cointegrating eqn(s) at the 0.05 level\x05\x05\x05\x05
* denotes rejection of the hypothesis at the 0.05 level\x05\x05\x05\x05
**MacKinnon-Haug-Michelis (1999) p-values\x05\x05\x05\x05
\x05\x05\x05\x05
Unrestricted Cointegration Rank Test (Maximum Eigenvalue)\x05\x05\x05\x05
\x05\x05\x05\x05
Hypothesized\x05\x05Max-Eigen\x050.05\x05
No.of CE(s)\x05Eigenvalue\x05Statistic\x05Critical Value\x05Prob.**
\x05\x05\x05\x05
None *\x05 0.664885\x05 24.05217\x05 21.13162\x05 0.0188
At most 1\x05 0.470765\x05 13.99910\x05 14.26460\x05 0.0550
At most 2\x05 0.143056\x05 3.396423\x05 3.841466\x05 0.0653
\x05\x05\x05\x05
Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level\x05\x05\x05\x05
* denotes rejection of the hypothesis at the 0.05 level\x05\x05\x05\x05
**MacKinnon-Haug-Michelis (1999) p-values\x05\x05\x05\x05
\x05\x05\x05\x05
Unrestricted Cointegrating Coefficients (normalized by b'*S11*b=I):\x05\x05\x05\x05
\x05\x05\x05\x05
LNTFP\x05LNOFDI\x05LNS\x05\x05
30.05579\x05-2.264232\x05-2.547612\x05\x05
12.11055\x05 2.298884\x05-2.846867\x05\x05
-2.975400\x05 3.045426\x05-2.971431\x05\x05
\x05\x05\x05\x05
\x05\x05\x05\x05
Unrestricted Adjustment Coefficients (alpha):\x05\x05\x05\x05
\x05\x05\x05\x05
D(LNTFP)\x05-0.022466\x05 0.002338\x05-0.004361\x05
D(LNOFDI)\x05-0.002442\x05-0.073448\x05-0.026211\x05
D(LNS)\x05-0.005357\x05-0.014135\x05 0.013305\x05
\x05\x05\x05\x05
\x05\x05\x05\x05
1 Cointegrating Equation(s):\x05\x05Log likelihood\x05 111.7347\x05
\x05\x05\x05\x05
Normalized cointegrating coefficients (standard error in parentheses)\x05\x05\x05\x05
LNTFP\x05LNOFDI\x05LNS\x05\x05
1.000000\x05-0.075334\x05-0.084763\x05\x05
\x05 (0.02389)\x05 (0.02110)\x05\x05
\x05\x05\x05\x05
Adjustment coefficients (standard error in parentheses)\x05\x05\x05\x05
D(LNTFP)\x05-0.675241\x05\x05\x05
\x05 (0.14561)\x05\x05\x05
D(LNOFDI)\x05-0.073407\x05\x05\x05
\x05 (0.92967)\x05\x05\x05
D(LNS)\x05-0.161007\x05\x05\x05
\x05 (0.29846)\x05\x05\x05
\x05\x05\x05\x05
\x05\x05\x05\x05
2 Cointegrating Equation(s):\x05\x05Log likelihood\x05 118.7343\x05
\x05\x05\x05\x05
Normalized cointegrating coefficients (standard error in parentheses)\x05\x05\x05\x05
LNTFP\x05LNOFDI\x05LNS\x05\x05
1.000000\x05 0.000000\x05-0.127467\x05\x05
\x05\x05 (0.00843)\x05\x05
0.000000\x05 1.000000\x05-0.566869\x05\x05
\x05\x05 (0.09582)\x05\x05
\x05\x05\x05\x05
Adjustment coefficients (standard error in parentheses)\x05\x05\x05\x05
D(LNTFP)\x05-0.646922\x05 0.056245\x05\x05
\x05 (0.15590)\x05 (0.01552)\x05\x05
D(LNOFDI)\x05-0.962901\x05-0.163318\x05\x05
\x05 (0.81939)\x05 (0.08159)\x05\x05
D(LNS)\x05-0.332184\x05-0.020364\x05\x05
\x05 (0.30199)\x05 (0.03007)

三变量误差修正模型：两个解释变量,用JJ协整检验后得到以下结果,如何看这个结果?怎样做误差修正模型?Date:01/13/12 Time:12:06\x05\x05\x05\x05Sample (adjusted):1989 2010\x05\x05\x05\x05Included observations:22 afte
Trace test indicates 2 cointegrating eqn(s) at the 0.05 level
Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level
这两句说明了之间确定存在协整关系,trace检验结果是两个协整向量,max检验结果是一个协整向量
要是只是为了做vecm 那么下面的协整方程不需要仔细研究
因为vecm就是有协整约束的var模型,cointegration检验通过的话就直接选择做vecm就可以了.在var type里选择