VAR里面做diagnostics test（包括自相关，异方差，normality）,结果如何看？

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

用VAR里面的residual test，有autocorrelation LM test, 怀特检验和normality test，不明白结果如何分别判断出是不是会有自相关，异方差和normality。结果如下：
Sample: 1985Q1 2010Q3  
Included observations: 99  
  
Lags LM-Stat Prob
  
1  12.55659  0.7049
2  13.48256  0.6372
3  19.88127  0.2256
  
Probs from chi-square with 16 df.  



VEC Residual Heteroskedasticity Tests: Includes Cross Terms     
Date: 09/03/11   Time: 18:50     
Sample: 1985Q1 2010Q3     
Included observations: 99     
     
     
   Joint test:     
     
Chi-sq df Prob.   
     
637.0631 540  0.0025   
     
     
   Individual components:     
     
Dependent R-squared F(54,44) Prob. Chi-sq(54) Prob.
     
res1*res1  0.644706  1.478541  0.0915  63.82592  0.1693
res2*res2  0.523146  0.893916  0.6550  51.79148  0.5600
res3*res3  0.482004  0.758198  0.8344  47.71838  0.7139
res4*res4  0.818837  3.682883  0.0000  81.06490  0.0100
res2*res1  0.784525  2.966673  0.0002  77.66801  0.0191
res3*res1  0.659504  1.578206  0.0603  65.29086  0.1396
res3*res2  0.615377  1.303664  0.1832  60.92237  0.2410
res4*res1  0.836105  4.156752  0.0000  82.77440  0.0071
res4*res2  0.807198  3.411349  0.0000  79.91256  0.0125
res4*res3  0.727588  2.176300  0.0044  72.03124  0.0510
     
     



VEC Residual Normality Tests   
Orthogonalization: Residual Correlation (Doornik-Hansen)   
Null Hypothesis: residuals are multivariate normal   
Date: 09/03/11   Time: 18:52   
Sample: 1985Q1 2010Q3   
Included observations: 99   
   
   
Component Skewness Chi-sq df Prob.
   
1  0.159301  0.470085 1  0.4929
2  0.033848  0.021429 1  0.8836
3 -0.431809  3.253826 1  0.0713
4 -0.855223  10.87639 1  0.0010
   
Joint   14.62173 4  0.0056
   
   
Component Kurtosis Chi-sq df Prob.
   
1  2.170197  3.714644 1  0.0539
2  2.068698  4.433743 1  0.0352
3  2.536771  3.371966 1  0.0663
4  4.259721  0.282659 1  0.5950
   
Joint   11.80301 4  0.0189
   
   
Component Jarque-Bera df Prob.
   
1  4.184729 2  0.1234
2  4.455172 2  0.1078
3  6.625792 2  0.0364
4  11.15905 2  0.0038
   
Joint  26.42474 8  0.0009
   
   
拜托拜托，论文真的是急用，不知道有木有高手可以指导一下~~

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：diagnostics Diagnostic normality Agnostic normal Cross 如何 怀特

楼主的问题后来解决了吗？我也碰到同样的问题，求助！！！！