字号：大 中 小

1、随机效应模型情况下，如果有变量的参数不显著时，hausman检验有效不，以下是随机效应的结果：
Dependent Variable: LOG(EX?)   
Method: Pooled EGLS (Cross-section random effects)   
Date: 03/16/10   Time: 15:34   
Sample: 1996 2008   
Included observations: 13   
Cross-sections included: 27   
Total pool (balanced) observations: 351   
Swamy and Arora estimator of component variances   
   
Variable Coefficient Std. Error t-Statistic Prob.  
   
C -33.66431 4.754878 -7.079952 0.0000
LOG(RJGDP?) 1.036394 0.183526 5.647111 0.0000
LOG(SCPCL?) 4.973280 0.453128 10.97544 0.0000
LOG(DIS?) -1.581683 0.408873 -3.868395 0.0001
D2? -1.003344 0.349901 -2.867509 0.0044
Random Effects (Cross)   
JAPAN--C 1.500218   
USA--C 4.249477   
REPKOREA--C 0.094872   
GERMANY--C 1.431548   
UK--C 0.617836   
SPAIN--C 1.246089   
CANADA--C 2.044952   
NERTHERLANDS--C -0.036543   
FRANCE--C 0.414250   
RUSSIAN--C 1.367327   
POLAND--C 0.376017   
PORTUGAL--C -0.488071   
THAILAND--C 0.628750   
MALAYSIA--C 0.471181   
PHILIPPINES--C 0.284563   
ITALY--C -1.062721   
SWEDEN--C -1.563098   
DENMARK--C -2.122384   
AUSTRALIA--C -0.301839   
INDONESIA--C 1.066160   
ISRAEL--C -3.001967   
SINGAPORE--C -1.338435   
NEWZEALAND--C -1.401194   
GREECE--C -1.796534   
NORWAY--C -3.285261   
MEXICO--C 0.889711   
VIETNAM--C -0.284902   
   
Effects Specification   
   S.D.   Rho  
   
Cross-section random   1.506125 0.7606
Idiosyncratic random   0.844925 0.2394
   
Weighted Statistics   
   
R-squared 0.557616     Mean dependent var  0.411192
Adjusted R-squared 0.552502     S.D. dependent var  1.280519
S.E. of regression 0.856607     Sum squared resid  253.8862
F-statistic 109.0315     Durbin-Watson stat  0.647789
Prob(F-statistic) 0.000000   
   
Unweighted Statistics   
   
R-squared 0.295889     Mean dependent var  2.674569
Sum squared resid 1166.856     Durbin-Watson stat  0.140947

2、PANAL DATA中DW值与意义不