求助关于collinearity的问题

tanshai

3414

收藏 2016-05-15

面板数据是11个公司从1935年到1954年的数据，大概是这样

然后是要求对这11个公司的inv,v,k分别求平均值，并做between regression
我是这么做的

tabulate year, gen(g)

reg inv g1 g2 g3 g4 g5 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20, nocons

predict inv_bar

reg v g1 g2 g3 g4 g5 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20, nocons

predict v_bar

reg k g1 g2 g3 g4 g5 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20, nocons

predict k_bar

xtreg inv_bar v_bar k_bar, be

但是结果显示

请问是什么原因导致的呢？正确的做法是什么呢？

万分感谢！

附件列表

1.jpg

原图尺寸 223.75 KB

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

tanshai

2016-5-16 05:14:26

我发现应该tabulate firm, gen(f)，但是为什么xtreg inv_bar v_bar k_bar（平均值）, be和xtreg inv v k, be得出的结果一样呢？
. xtreg inv_bar v_bar k_bar, be

Between regression (regression on group means)  Number of obs    =    220
Group variable: firm                         Number of groups =       11

R-sq:  within  =    .                      Obs per group: min =       20
   between = 0.8644                                     avg =    20.0
   overall = 0.8644                                     max =       20

                                             F(2,8)          =    25.50
sd(u_i + avg(e_i.))=  79.55147                Prob > F          = 0.0003

------------------------------------------------------------------------------
   inv_bar |    Coef. Std. Err.    t P>|t|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
   v_bar | .1345988 .0268845    5.01 0.001    .0726029 .1965946
   k_bar | .029688 .1746056    0.17 0.869 -.3729532 .4323292
   _cons |  -7.382482 40.44366 -0.18 0.860 -100.6457 85.88077
------------------------------------------------------------------------------

. estimates store BEreg

. xtreg inv v k, be

Between regression (regression on group means)  Number of obs    =    220
Group variable: firm                         Number of groups =       11

R-sq:  within  = 0.4706                      Obs per group: min =       20
   between = 0.8644                                     avg =    20.0
   overall = 0.7624                                     max =       20

                                             F(2,8)          =    25.50
sd(u_i + avg(e_i.))=  79.55147                Prob > F          = 0.0003

------------------------------------------------------------------------------
      inv |    Coef. Std. Err.    t P>|t|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
         v | .1345988 .0268845    5.01 0.001    .0726029 .1965946
         k | .029688 .1746056    0.17 0.869 -.3729532 .4323292
   _cons |  -7.382483 40.44366 -0.18 0.860 -100.6457 85.88077
------------------------------------------------------------------------------