全部版块 我的主页
论坛 计量经济学与统计论坛 五区 计量经济学与统计软件 Stata专版
4194 2
2007-10-14

sum it it1 dt1 gt1 cft

VariableObsMeanStd. Dev.MinMax
it1998.0832613.134807-.3837973.8216538
it11998.0946333.1402503-.3837973.8721095
dt11998.4643617.1680077.0684844.9001655
gt11998-.0054551.6564586-1.5555642.517123
cft1998.0662256.0775479-.1810507.3600312

 regress it it1 dt1 gt1 cft

      Source |       SS       df       MS              Number of obs =    1998
-------------+------------------------------           F(  4,  1993) =   71.71
       Model |  4.56579675     4  1.14144919           Prob > F      =  0.0000
    Residual |  31.7255411  1993  .015918485           R-squared     =  0.1258
-------------+------------------------------           Adj R-squared =  0.1241
       Total |  36.2913378  1997  .018172928           Root MSE      =  .12617

------------------------------------------------------------------------------
          it |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         it1 |   .1687116   .0234238     7.20   0.000      .122774    .2146492
         dt1 |   -.082543   .0168539    -4.90   0.000    -.1155961     -.04949
         gt1 |   .0233332   .0050948     4.58   0.000     .0133415    .0333248
         cft |    .319183   .0373506     8.55   0.000     .2459327    .3924333
       _cons |   .0846146    .009103     9.30   0.000     .0667621     .102467
-------------------------------------------------------------

xi:xtreg it it1 dt1 gt1 cft i.year,fe
i.year            _Iyear_2001-2006    (naturally coded; _Iyear_2001 omitted)

Fixed-effects (within) regression               Number of obs      =      1998
Group variable: company                         Number of groups   =       333

R-sq:  within  = 0.1133                         Obs per group: min =         6
       between = 0.0934                                        avg =       6.0
       overall = 0.0000                                        max =         6

                                                F(9,1656)          =     23.51
corr(u_i, Xb)  = -0.6320                        Prob > F           =    0.0000

------------------------------------------------------------------------------
          it |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         it1 |   .0541041   .0263116     2.06   0.040     .0024965    .1057117
         dt1 |  -.3127219   .0347334    -9.00   0.000    -.3808478    -.244596
         gt1 |  -.0702787    .010142    -6.93   0.000    -.0901713   -.0503862
         cft |   .2236034   .0419513     5.33   0.000     .1413202    .3058865
 _Iyear_2002 |  -.0037339   .0090098    -0.41   0.679    -.0214057    .0139378
 _Iyear_2003 |  -.0064442   .0091742    -0.70   0.483    -.0244384      .01155
 _Iyear_2004 |   .0093567   .0094417     0.99   0.322    -.0091622    .0278757
 _Iyear_2005 |  -.0002629    .009849    -0.03   0.979    -.0195808    .0190549
 _Iyear_2006 |    .000966   .0101313     0.10   0.924    -.0189054    .0208375
       _cons |   .2081854   .0162801    12.79   0.000     .1762536    .2401172
-------------+----------------------------------------------------------------
     sigma_u |  .10917924
     sigma_e |  .11534182
         rho |  .47257298   (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0:     F(332, 1656) =     2.10           Prob > F = 0.0000

.
.
.
. xi:xtreg it it1 dt1 gt1 cft i.year,re
i.year            _Iyear_2001-2006    (naturally coded; _Iyear_2001 omitted)

Random-effects GLS regression                   Number of obs      =      1998
Group variable: company                         Number of groups   =       333

R-sq:  within  = 0.0109                         Obs per group: min =         6
       between = 0.5767                                        avg =       6.0
       overall = 0.1370                                        max =         6

Random effects u_i ~ Gaussian                   Wald chi2(9)       =    315.66
corr(u_i, X)       = 0 (assumed)                Prob > chi2        =    0.0000

------------------------------------------------------------------------------
          it |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         it1 |   .1437765   .0240017     5.99   0.000     .0967339     .190819
         dt1 |  -.0657266   .0171307    -3.84   0.000    -.0993022   -.0321511
         gt1 |   .0284564   .0051972     5.48   0.000       .01827    .0386428
         cft |   .3220453   .0372327     8.65   0.000     .2490705      .39502
 _Iyear_2002 |  -.0086856   .0097604    -0.89   0.374    -.0278156    .0104443
 _Iyear_2003 |  -.0227694   .0098113    -2.32   0.020    -.0419993   -.0035396
 _Iyear_2004 |  -.0143972   .0099181    -1.45   0.147    -.0338363    .0050419
 _Iyear_2005 |   -.040167   .0100166    -4.01   0.000    -.0597993   -.0205347
 _Iyear_2006 |   -.039464   .0101472    -3.89   0.000     -.059352   -.0195759
       _cons |   .0999176   .0107153     9.32   0.000     .0789159    .1209193
-------------+----------------------------------------------------------------
     sigma_u |          0
     sigma_e |  .11534182
         rho |          0   (fraction of variance due to u_i)
------------------------------------------------------------------------------

. xttest0

Breusch and Pagan Lagrangian multiplier test for random effects

        it[company,t] = Xb + u[company] + e[company,t]

        Estimated results:
                         |       Var     sd = sqrt(Var)
                ---------+-----------------------------
                      it |   .0181729        .134807
                       e |   .0133037       .1153418
                       u |          0              0

        Test:   Var(u) = 0
                              chi2(1) =     1.92
                          Prob > chi2 =     0.1656
qui xtreg it it1 dt1 gt1 cft,fe     
estimates store fe
qui xtreg it it1 dt1 gt1 cft,re   
estimates store re
hausman fe
     

  
                 ---- Coefficients ----
             |      (b)          (B)            (b-B)     sqrt(diag(V_b-V_B))
             |       fe           re         Difference          S.E.
-------------+----------------------------------------------------------------
         it1 |    .0486486     .1687116        -.120063        .0086528
         dt1 |   -.3081782     -.082543       -.2256351        .0272443
         gt1 |    -.068016     .0233332       -.0913492         .008058
         cft |    .2220242      .319183       -.0971588        .0188305
------------------------------------------------------------------------------
                           b = consistent under Ho and Ha; obtained from xtreg
            B = inconsistent under Ha, efficient under Ho; obtained from xtreg

    Test:  Ho:  difference in coefficients not systematic

                  chi2(4) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                          =     1443.37
                Prob>chi2 =      0.0000

这里调整R2如此小,问题出在哪里呢?恳求高手指点!非常感谢!

[此贴子已经被作者于2007-10-15 0:34:00编辑过]

二维码

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

全部回复
2007-10-15 14:57:00
面板数据的调整R方好像都不大,我看F和P合理,hausman的chi2(4) 显著,结果可以用啊.
二维码

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

2010-3-20 00:37:31
同意楼上的说法,在微观面板数据里 ,R^2不说明问题
二维码

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

相关推荐
栏目导航
热门文章
推荐文章

说点什么

分享

扫码加好友,拉您进群
各岗位、行业、专业交流群