Pool unit root test: Summary  
Date: 07/09/08   Time: 06:56 
Sample: 2005M01 2007M12  
Series: DR_QZHONGXING, DR_QDONGBEI, DR_QANGANG,
        DR_QHUANENG, DR_QZHONGHAI, DR_QZHONGSHIHUA,
        DR_QNANFANG, DR_QGUANGZHOU, DR_QJIANGXI,
        DR_QSHENGAOSU, DR_QHAILUO, DR_QGUANGCHUAN,
        DR_QMAGANG, DR_QBEIREN, DR_QCHUANGYE,
        DR_QYIHUA, DR_QJINGWEI 
Exogenous variables: Individual effects
Automatic selection of maximum lags 
Automatic selection of lags based on SIC: 0
Newey-West bandwidth selection using Bartlett kernel
Balanced observations for each test  
    
    
   Cross- 
Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process)
Levin, Lin & Chu t* -3.59312  0.0002  17  595
Breitung t-stat -2.53451  0.0056  17  578
    
Null: Unit root (assumes individual unit root process)
Im, Pesaran and Shin W-stat  -3.56874  0.0002  17  595
ADF - Fisher Chi-square  66.4208  0.0007  17  595
PP - Fisher Chi-square  62.5071  0.0021  17  595
    
Null: No unit root (assumes common unit root process)
Hadri Z-stat  6.27730  0.0000  17  612
    
    
** Probabilities for Fisher tests are computed using an asympotic Chi
        -square distribution. All other tests assume asymptotic normality.

从Levin, Lin & Chu t*和Breitung t-stat 上应该拒绝原假设，即各截面序列不存在共同的单位根过程。从Hadri Z-stat 检验上也是拒绝原假设，认为存在单位根。--------我想，合起来的意思是不是，虽然不存在共同的单位根过程，但是可能存在个体不同的单位根过程。但Im, Pesaran and Shin W-stat等另外三个检验也是拒绝原假设，认为各截面序列不存在个体不同的单位根过程。这到底是怎么回事呢？

[此贴子已经被作者于2008-7-9 7:48:12编辑过]