Time inequality analysis - Generating Gini Index & 90/10 pctile

[Andrea Ciaccio]

Hi everyone,
I am using Italian Time use data, I will post a fake example of it:

* Example generated by -dataex-. To install: ssc install dataex
clear
input str6 individual double(play_mother sport_mother)
"1"   30 20
"2"   0  0
"3"  40  30
"4"   0  10
"5"   70  0
"6"   0  0
"7"   100  0
"8"   0  0
"9"   0  20
"10"  150  90
end

where "individua"l is the identifier of individual "j", "play_mother" and "sport_mother" are variables indicating the amount of time spent with mother in playing and doing sport.
I would like to analyze the distribution of time spent in playing and doing sports by the child with the mother, in particular I would like to compute the Gini Index and the ratio between the 90th and 10th percentiles! Is there a way to compute them in Stata and, eventually, to export them (for example using something like "estout" "outreg")?
Thank you in advance
Andrea

[Stephen Jenkins]

clear
input str6 individual double(play_mother sport_mother)
"1"   30 20
"2"   0  0
"3"  40  30
"4"   0  10
"5"   70  0
"6"   0  0
"7"   100  0
"8"   0  0
"9"   0  20
"10"  150  90
end

ineqdec0 play_mother // ssc describe ineqdec0; ssc describe ineqdeco
ineqdec0 sport_mother
return list

results in

. clear

. input str6 individual double(play_mother sport_mother)

     individ~l  play_mot~r  sport_mo~r
  1. "1"   30 20
  2. "2"   0  0
  3. "3"  40  30
  4. "4"   0  10
  5. "5"   70  0
  6. "6"   0  0
  7. "7"   100  0
  8. "8"   0  0
  9. "9"   0  20
 10. "10"  150  90
 11. end

.
. ineqdec0 play_mother // ssc describe ineqdec0; ssc describe ineqdeco
 
Warning: play_mother has 5 values = 0. Used in calculations
Note: p25 (and smaller percentiles) <= 0
 
Percentile ratios

----------------------------------------------------------
  All obs |    p90/p10     p90/p50     p10/p50     p75/p25
----------+-----------------------------------------------
          |                  8.333       0.000           
----------------------------------------------------------
 
Generalized Entropy index GE(2), and Gini coefficient

----------------------------------
  All obs |      GE(2)        Gini
----------+-----------------------
          |    0.81164     0.65385
----------------------------------

. ineqdec0 sport_mother
 
Warning: sport_mother has 5 values = 0. Used in calculations
Note: p25 (and smaller percentiles) <= 0
 
Percentile ratios

----------------------------------------------------------
  All obs |    p90/p10     p90/p50     p10/p50     p75/p25
----------+-----------------------------------------------
          |                 12.000       0.000           
----------------------------------------------------------
 
Generalized Entropy index GE(2), and Gini coefficient

----------------------------------
  All obs |      GE(2)        Gini
----------+-----------------------
          |    1.21280     0.70000
----------------------------------

.. return list

scalars:
              r(wgini) =  5.100000000000001
                r(ge2) =  1.21280276816609
               r(gini) =  .7
             r(p75p50) =  4
             r(p25p50) =  0
             r(p10p50) =  0
             r(p90p50) =  12
             r(p75p25) =  .
             r(p90p10) =  .
                r(p95) =  90
                r(p90) =  60
                r(p75) =  20
                r(p50) =  5
                r(p25) =  0
                r(p10) =  0
                 r(p5) =  0
                r(max) =  90
                r(min) =  0
                  r(N) =  10
               r(sumw) =  10
                 r(sd) =  27.9085809185793
                r(Var) =  778.8888888888889
               r(mean) =  17

You can manipulate the results stored in r(), e.g. using Ben Jann's esto

[Andrea Ciaccio]

Thank You very much Stephen!

[Andrea Ciaccio]

Dear Professor Jenkins, I tried with estout but it tells me that there is no estimation results stored...do I need to put other commands (like estpost) before proceeding with estout?

[Stephen Jenkins]

When I wrote

You can manipulate the results stored in r(), e.g. using Ben Jann's esto

, I should have written

You can manipulate the results stored in r(), e.g. using Ben Jann's estadd

. Look at the help file for it, and see how it will place in e()

[Andrea Ciaccio]

Thank you very much for your help!