univariate test

[ALKEBSEE RADWAN]

Hello everyone,
I have a question about how to apply uni-variate test in stata.
My model as follows:

reg log_patent log_CEOPAY firmsize leverage board_indp log_RD est_firm region_status num_of_ceo_major_expertise num_of_id_financial_expertise num_of_ceo_degree_expert num_of_ceo_job_experience n_of_fem_ceo av_ceo_shareholding av_ceo_age big4 soe btm1 tobinq1 roa log_develpo_expens

which is
how to compare between the means of two samples (in my model the variable

soe

)

i thought i can use Ttest but this test use only for one variable.
i want to get a table like the one in the attachment

for more detail please see the picture attached
thank you in advance

[Carlo Lazzaro]

Alkebsee:
univariate in the (deprecated) screenshot you posted means that a -ttest- was conducted on each variable separately, setting aside the correlation between them (unlike what you get from -regress-)..
That said, for each variable you can challege yourself with something along the following lines:

. set obs 100
number of observations (_N) was 0, now 100

. g CSR=runiform()*1000

. g firm=0 in 1/50
(50 missing values generated)

. replace firm=1 if missing(firm)
(50 real changes made)

. label define firm 0 "private" 1 "public"

. label val firm firm

. ttest CSR, by(firm) unequal

Two-sample t test with unequal variances
------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
 private |      50     544.965    43.77794    309.5568    456.9899      632.94
  public |      50    449.7129    42.91137    303.4292    363.4792    535.9465
---------+--------------------------------------------------------------------
combined |     100    497.3389      30.869      308.69    436.0881    558.5897
---------+--------------------------------------------------------------------
    diff |             95.2521    61.30166               -26.39966    216.9039
------------------------------------------------------------------------------
    diff = mean(private) - mean(public)                           t =   1.5538
Ho: diff = 0                     Satterthwaite's degrees of freedom =  97.9609

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.9383         Pr(|T| > |t|) = 0.1235          Pr(T > t) = 0.0617

.

Eventually, you can think you can automatize the whole stuff via -foreach-.

[ALKEBSEE RADWAN]

@Carlo Lazzaro thank you so much. I got it.

Best Regards
Alkebsee