我作毕业论文时遇到了这个问题,用eviews几个单位根检验得出的结果完全不同,那这算不算有单位根呢

 Panel unit root test: Summary     
Date: 08/27/08   Time: 12:24    
Sample: 1995 2005    
Exogenous variables: Individual effects, individual linear trends    
User specified maximum lags    
Automatic selection of lags based on SIC: 0 to 2    
Newey-West bandwidth selection using Bartlett kernel    
    
   Cross- 
Method                                             Statistic                 Prob.** sections Obs
Null: Unit root (assumes common unit root process)     
Levin, Lin & Chu  t*                             -35.8379            0.0000  30  201
Breitung t-stat                                      5.59752             1.0000  30  171
    
Null: Unit root (assumes individual unit root process)     
Im, Pesaran and Shin W-stat                -11.2128            0.0000  30  201
ADF - Fisher Chi-square                     145.893              0.0000  30  201
PP - Fisher Chi-square                        268.372               0.0000  30  240
    
Null: No unit root (assumes common unit root process)     
Hadri Z-stat                                          30.7403                 0.0000  30  270
    
** Probabilities for Fisher tests are computed using an asympotic Chi    
        -square distribution. All other tests assume asymptotic normality.    

也就是除了Breitung和Hadri外,都拒绝有单位根,那这算不算有单位根啊

[此贴子已经被作者于2008-8-27 12:27:29编辑过]