_majia_ 发表于 2012-6-15 21:56 
先试AR 2
Conditional Least Squares Estimation
Standard Approx
Parameter Estimate Error t Value Pr > |t| Lag
MU 1.45724 1.09670 1.33 0.1868 0
MA1,1 0.69736 0.11260 6.19 <.0001 12
AR1,1 -0.10807 0.09818 -1.10 0.2735 2
AR1,2 0.0096408 0.15654 0.06 0.9510 12
Constant Estimate 1.600683
Variance Estimate 1040.589
Std Error Estimate 32.25817
AIC 1060.749
SBC 1071.478
Number of Residuals 108
* AIC and SBC do not include log determinant.
Correlations of Parameter Estimates
Parameter MU MA1,1 AR1,1 AR1,2
MU 1.000 -0.081 0.007 -0.072
MA1,1 -0.081 1.000 -0.017 0.739
AR1,1 0.007 -0.017 1.000 -0.083
AR1,2 -0.072 0.739 -0.083 1.000
Autocorrelation Check of Residuals
To Chi- Pr >
Lag Square DF ChiSq --------------------Autocorrelations--------------------
6 4.55 3 0.2075 -0.170 -0.005 0.012 -0.052 0.088 -0.034
12 6.32 9 0.7078 0.014 0.060 0.021 0.102 0.004 -0.004
18 10.55 15 0.7841 -0.035 -0.059 0.067 -0.152 0.004 -0.025
24 13.20 21 0.9014 -0.036 -0.020 -0.087 0.017 -0.051 -0.084