Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Missing F statistics.


From   natasha agarwal <agarwana2@googlemail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Missing F statistics.
Date   Tue, 22 Jun 2010 12:46:39 +0100

Dear Everyone,

I am trying to estimate an augmented production using a within
estimator on an unbalanced panel data. Panel Identifier in my case is
a firm.

When I cluster at the industry level, I get a missing F statistics.

where fs1fs10 is an interaction term of FDI in that industry with an
industry dummy (lnvfdi*i1). The same is done for all the 10 industries
in the dataset

xtreg lnrval lnk lnw lnvfdi fs1 fs3-fs10 y14-y18, fe vce(cluster indm)

Fixed-effects (within) regression               Number of obs      =      6147
Group variable: number                          Number of groups   =      1609

R-sq:  within  = 0.5130                         Obs per group: min =         1
       between = 0.2057                                        avg =       3.8
       overall = 0.2103                                        max =         5

                                                F(6,9)             =         .
corr(u_i, Xb)  = -0.7352                        Prob > F           =         .

                                  (Std. Err. adjusted for 10 clusters in indm)
------------------------------------------------------------------------------
             |               Robust
      lnrval |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         lnk |   .2022405   .0217397     9.30   0.000     .1530619     .251419
         lnw |   .7147967   .0178031    40.15   0.000     .6745233    .7550701
      lnvfdi |   .1595684   .0414696     3.85   0.004     .0657577    .2533791
         fs1 |   -.500915   .0377129   -13.28   0.000    -.5862277   -.4156024
         fs2 |  -.1892681     .02731    -6.93   0.000    -.2510477   -.1274886
         fs4 |  -.0979289   .0794079    -1.23   0.249     -.277562    .0817043
         fs5 |  -.1276782   .0345498    -3.70   0.005    -.2058353   -.0495212
         fs6 |  -.1352529   .0165971    -8.15   0.000    -.1727981   -.0977077
         fs7 |  -.2112656   .0191332   -11.04   0.000    -.2545478   -.1679833
         fs8 |   .2629032   .0671154     3.92   0.004     .1110776    .4147287
         fs9 |   .0308106   .0489407     0.63   0.545    -.0799009    .1415222
        fs10 |  -.2493683   .0194207   -12.84   0.000    -.2933011   -.2054355
         y14 |  -.1692296   .0617547    -2.74   0.023    -.3089283   -.0295308
         y15 |  -.0790967   .0418282    -1.89   0.091    -.1737186    .0155253
         y16 |   -.067962   .0334897    -2.03   0.073    -.1437211    .0077971
         y17 |  -.1063753   .0221839    -4.80   0.001    -.1565587   -.0561919
         y18 |  (dropped)
       _cons |  -.7143403   .6204304    -1.15   0.279    -2.117851    .6891708
-------------+----------------------------------------------------------------
     sigma_u |  1.4145653
     sigma_e |  .38918461
         rho |  .92963183   (fraction of variance due to u_i)
------------------------------------------------------------------------------

When I cluster it at the firm level, it works fine.



Fixed-effects (within) regression               Number of obs      =      6147
Group variable: number                          Number of groups   =      1609

R-sq:  within  = 0.5130                         Obs per group: min =         1
between = 0.2057                                        avg =       3.8
overall = 0.2103                                        max =         5

F(16,1608)         =    155.17
corr(u_i, Xb)  = -0.7352                        Prob > F           =    0.0000

(Std. Err. adjusted for 1609 clusters in number)

Robust
lnrval       Coef.   Std. Err.      t    P>t     [95% Conf. Interval]

lnk    .2022405   .0190487    10.62   0.000     .1648776    .2396033
lnw    .7147967   .0259381    27.56   0.000     .6639206    .7656728
lnvfdi    .1595684   .1151602     1.39   0.166    -.0663116    .3854483
fs1    -.500915   .3712889    -1.35   0.177    -1.229176    .2273461
fs2   -.1892681   .1189724    -1.59   0.112    -.4226254    .0440892
fs4   -.0979289   .1353951    -0.72   0.470    -.3634983    .1676406
fs5   -.1276782   .1347357    -0.95   0.343    -.3919543    .1365979
fs6   -.1352529    .117889    -1.15   0.251    -.3664851    .0959794
fs7   -.2112656   .1401514    -1.51   0.132    -.4861641     .063633
fs8    .2629032   .1889839     1.39   0.164    -.1077774    .6335838
fs9    .0308106    .153413     0.20   0.841    -.2700999    .3317212
fs10   -.2493683   .1234036    -2.02   0.043    -.4914171   -.0073195
y14   -.1692296    .036587    -4.63   0.000    -.2409928   -.0974664
y15   -.0790967    .026678    -2.96   0.003     -.131424   -.0267693
y16    -.067962   .0189171    -3.59   0.000    -.1050667   -.0308573
y17   -.1063753   .0154075    -6.90   0.000    -.1365962   -.0761544
y18   (dropped)
_cons   -.7143403   .4764399    -1.50   0.134    -1.648849    .2201682

sigma_u   1.4145653
sigma_e   .38918461
rho   .92963183   (fraction of variance due to u_i)

Can anyone please help me explain why is this happening?

Thanks
Natasha
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index