Creating a dummy variable - New User of Stata 15

[Keiran daniels]

Hello everyone, undergraduate Bsc Economics student here, I am struggling to use Stata 15. This may be a very simple question to those of you that are competent or experienced in using Stata, but it's has me confused.

I have a dataset (Argentinian enterprise panel survey) containing 527 variables from the years 2015,2016,2017 and 2018.

I am trying to

1. generate a new variable exp_dum (export dummy) which equals 1 if sales_exp is positive, otherwise 0.
2. generate new variable foreign_dum (foreign ownership dummy) which equals 1 if foreign is positive, otherwise 0.
3. generate new variable soe_dum (state ownership dummy) which equals 1 if soe is positive, otherwise 0.

Additionally, I want to produce a summary statistic table of all variables to include the three dummies variables. including observation number, mean, standard deviation, minimum, and maximum values etc

How would I go about generating these three dummy variables in stata 15?

Many thanks in advance

Kieran

[Nick Cox]

A general recipe for an indicator (you say dummy) that is 1 for positive values of a given variable, 0 for zero or negative values of the same and missing for missing ditto is

gen wanted  = myvar > 0 if myvar < .

See documentation at

https://www.stata.com/support/faqs/d...mmy-variables/

https://www.stata.com/support/faqs/d...rue-and-false/

https://www.stata-journal.com/articl...article=dm0099 (if you have access)

[Carlo Lazzaro]

Kieran (but seemingly subscribed as Keiran):
as an aside to Nick's helpful advice, you can create a table collecting all the descriptice statistics you're interested in via -tabstat-:

. use "C:\Program Files\Stata16\ado\base\a\auto.dta"
(1978 Automobile Data)

. tabstat foreign rep78, stat(count mean sd p50 min max)

   stats |   foreign     rep78
---------+--------------------
       N |        74        69
    mean |  .2972973  3.405797
      sd |  .4601885  .9899323
     p50 |         0         3
     min |         0         1
     max |         1         5
------------------------------

Moreover, if you're dealing with panel data, take a look at the -xt- suite in Stata .pdf manual.

[Keiran daniels]

Hello,

Thank you that has helped a lot. I hope you do not mind I have one more question if you are able to assist. The data I am using for my analysis is panel data, which command may I use for panel data if I want to produce a summary statistics table that contains the observation, mean, sd, min and max values of all variables including the new indicator variables I have now generated.

Kindest Regards

Keiran
(Stata 15)

[Nick Cox]

tabstat works fine with panel data. Did you try it?

[Carlo Lazzaro]

Keiran:
as Nick claim, you can easily -tabstat- with panel data, too:

se "C:\Program Files\Stata16\ado\base\a\auto.dta"
. tabstat age race , stat(count mean sd p50 min max)

   stats |       age      race
---------+--------------------
       N |     28510     28534
    mean |  29.04511  1.303392
      sd |  6.700584  .4822773
     p50 |        28         1
     min |        14         1
     max |        46         3
------------------------------

That said (but possibly off topic here), the -xt- suite provides other commands to explore some features of your data that may be good to know before challenging yourself with panel data regression:

. xtsum age

Variable         |      Mean   Std. Dev.       Min        Max |    Observations
-----------------+--------------------------------------------+----------------
age      overall |  29.04511   6.700584         14         46 |     N =   28510
         between |             5.485756         14         45 |     n =    4710
         within  |              5.16945   14.79511   43.79511 | T-bar = 6.05308

. xtsum race

Variable         |      Mean   Std. Dev.       Min        Max |    Observations
-----------------+--------------------------------------------+----------------
race     overall |  1.303392   .4822773          1          3 |     N =   28534
         between |             .4862111          1          3 |     n =    4711
         within  |                    0   1.303392   1.303392 | T-bar = 6.05689

.

[Keiran daniels]

Hello Nick & Carlo,

Thank you for the help, tabstat has worked fine, I am now using asdoc package to save the summary statistics table to a word document. However, is there a way I can reduce the values to 3 decimal places?

The command I am using is:

asdoc tabstat 'variables', stat(count mean sd p50 min max), save(ICC)

This method works, but I would like to reduce the values on the table to 3 decimals.

kindest regards

Keiran

[Nick Cox]

I've never used asdoc (which you should please explain as community-contributed). Attaullah Shah, the program author, or one of its users is likely to help.