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 on April 23, and its replacement, statalist.org is already up and running.


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

st: fixed effects with time-invariant variable and interaction term


From   Ferdinand Eibl <ferdinand.eibl@politics.ox.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: fixed effects with time-invariant variable and interaction term
Date   Tue, 30 Oct 2012 08:42:40 +0000

Dear Statalist,

We have estimated a Lagged Dependent Variable (LDV) model with fixed effects for authoritarian regimes (with the following equation: Yit=β0 + β1Yit-1+ β2Growth+ β3Regime Type*Growth+ β4Xit+Ni+ εit). The model tries to estimate the effect of growth on social spending p.c. in different autocratic regimes. In the estimation, we used interaction terms between regime types and economic growth and omitted the different regime types from the equation as they are perfectly collinear with the regime fixed effects. 

As we now want to interpret our results, we were wondering how to calculate the marginal effects. More generally, we were also unsure whether, from our results table below , we could conclude that there is a significant difference between different regime types. Please note that party regime is the reference category. Any kind of recommendations and hints would be highly appreciated. 

Kind regards,
Ferdinand

STATA COMMAND
xi: xtpcse welfare_pc lagwelfare_pc  monarchy_growth personal_growth  military_growth ln_cgdp GDPgrowthpc  openc debtservice  direct_tax_gdp   ln_aid_pc  ln_rents_pc  urbanpop   dependency  polity  deficit_gdp imf  defence_pc gwf_duration i.regime, pairwise correlation(psar1)

RESULTS
------------------------------------------------------------------------------
             |                            Panel-corrected 
  welfare_pc |             Coef.   Std. Err.      z    P>|z|     [90% Conf. Interval]
-------------+----------------------------------------------------------------
lagwelf~e_pc |           .5927318   .1177409     5.03   0.000     .3990653    .7863982
monarchy_growthh |  -196.9249   70.88603    -2.78   0.005    -313.5221   -80.32778
personal_growthh |    -112.8295     58.188    -1.94   0.052    -208.5402   -17.11872
military_growth |       -175.469   69.33548    -2.53   0.011    -289.5157   -61.42223
ln_cgdp |                    33.54026   30.56969     1.10   0.273    -16.74241    83.82293
growth |                     173.2595   65.48369     2.65   0.008     65.54841    280.9706
openc |                       .6272041   .1929411     3.25   0.001     .3098441     .944564
debtservice |              .2663759   .6796557     0.39   0.695    -.8515582     1.38431
direct_tax~p |            -5.630332   262.5316    -0.02   0.983    -437.4564    426.1958
ln_aid_pc |                 3.78063   4.304718     0.88   0.380         -3.3    10.86126
ln_rents_pc |              -2.575868   1.401622    -1.84   0.066    -4.881331   -.2704041
urbanpop |                 3.645545   3.535471     1.03   0.302    -2.169788    9.460878
dependency |             -8.524525   2.860049    -2.98   0.003    -13.22889   -3.820163
polity |                       -1.634748   .9341539    -1.75   0.080    -3.171294   -.0982013
deficit_gdp |               -32.23502    57.8288    -0.56   0.577    -127.3549    62.88489
imf |                           -4.356364   3.477033    -1.25   0.210    -10.07557    1.362846
defence_pc |              .8791647   .1779755     4.94   0.000      .586421    1.171908
gwf_duration |           -1.545701   1.846139    -0.84   0.402     -4.58233    1.490927


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


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