Hi all,
After reading many many topics I have a question on the possibilities of any test to check whether the coefficients differ significantly or not.
I know about the -test- or -testparm- command but unfortunately can't find what I am looking for.
I test the effect of CEO pay scheme dependence on shareholder return (1=no, 2=yes) on the number of relative layoffs of the firm (percentage). Using paneldata, 5500 firms over 12 year:.
I want to split the effect of my main dependent for firms that made more or less profit than the year before, therefore a binairy variable (1= decreasing, 0=increasing).
xtreg RLO_per CeoPay#D_prof logsize logemployee tobinsq lev prof Domestic BoardSize i.n_ShhlType i.year i.n_Industry i.n_Country, vce(robust)
Now I want to test whether the coefficient of CeoPay=yes given decreasing profits (=0.2976) differs from the coefficient CeoPay=no given decreasing profits (=0.0683).
Anyone knows a proper method and stata command to do so?
Kinds regards,
Stan
Random-effects GLS regression Number of obs = 39,140
Group variable: ID Number of groups = 5,531
R-sq: Obs per group:
within = 0.0169 min = 1
between = 0.1029 avg = 7.1
overall = 0.0399 max = 14
Wald chi2(148) = .
corr(u_i, X) = 0 (assumed) Prob > chi2 = .
(Std. Err. adjusted for 5,531 clusters in ID)
---------------------------------------------------------------------------------------------------------------------------
| Robust
RLO_per | Coef. Std. Err. z P>|z| [95% Conf. Interval]
----------------------------------------------------------+----------------------------------------------------------------
CeoPay#D_prof |
1 1 | .0683347 .0302017 2.26 0.024 .0091404 .1275289
2 0 | .0298435 .0413767 0.72 0.471 -.0512534 .1109404
2 1 | .2976375 .0570798 5.21 0.000 .1857632 .4095118
|
logsize | .1262604 .0180784 6.98 0.000 .0908274 .1616933
logemployee | .0596924 .0114406 5.22 0.000 .0372692 .0821157
tobinsq | .0300837 .0107114 2.81 0.005 .0090897 .0510778
lev | .0175412 .1009887 0.17 0.862 -.180393 .2154755
prof | -2.063631 .2746226 -7.51 0.000 -2.601882 -1.525381
Domestic | -.023963 .0411957 -0.58 0.561 -.1047052 .0567792
BoardSize | .0036602 .0058319 0.63 0.530 -.0077701 .0150906
Related Posts with Wald test or any related for regression with interaction term
Exporting confidence intervals that loop through all vars in an excel formatFor each of my regional data files, I am trying to export confidence interval tables in a csv file. …
Why the enhanced module of lclogit (lclogit2) does not converge?Dears I am estimating latent class models and i have some problems with the command lclogit2. Appar…
Insignificant F statistics, R square approximately equal to 0. Insignficant t statisticsDear Stata list members Advance apologies for asking a naive question. I conducted an analysis to c…
Percentage in bar graphHello everyone. I am trying to create a bar chart which evaluates the labor market participation by…
how to make a descriptive statistics like the picture (including Jarque‐Bera test for normality) and output the result"sum,de" did not include Jarque‐Bera test for normality, and how to use the panel data to achieve th…
Subscribe to:
Post Comments (Atom)
0 Response to Wald test or any related for regression with interaction term
Post a Comment