Dependent Variable: LOGY?   
Method: Pooled EGLS (Cross-section weights)   
Date: 05/28/12   Time: 17:15   
Sample: 1984 2003   
Included observations: 20   
Cross-sections included: 13   
Total pool (balanced) observations: 260   
Linear estimation after one-step weighting matrix   
   
Variable Coefficient Std. Error t-Statistic Prob.  
   
C 10.00899 0.567415 17.63963 0.0000
LOGG? -0.353696 0.147584 -2.396577 0.0173
LOGL? 0.073199 0.022989 3.184167 0.0016
LOGP? 0.028593 0.088851 0.321811 0.7479
LOGM? -0.072772 0.022831 -3.187409 0.0016
   
Weighted Statistics   
   
R-squared 0.078970     Mean dependent var  11.36641
Adjusted R-squared 0.064522     S.D. dependent var  4.518703
S.E. of regression 0.238946     Sum squared resid  14.55926
F-statistic 5.465967     Durbin-Watson stat  0.183451
Prob(F-statistic) 0.000308   
   
Unweighted Statistics   
   
R-squared 0.098324     Mean dependent var  8.414076
Sum squared resid 16.66519     Durbin-Watson stat  0.128985
结果发现拟合度和DW值都很小，如果用固定效应和随机效应的话系数符号和预测的相反 请各位大侠帮忙分析下问题可能出现在哪里 不胜感激