Dear Stata users,

I have an interesting finding regards to twoway graph's legend option, put it precisely, it's about label option of legend. I use -lorenz- command (SSC) to estimate wage inequality between subgroups of union membership. I get a matrix named e(G) which stores gini indexes of union and nonunion groups. Then I use lorenz graph to get corresponding Gini curve. In the graph's legend options, I typed this code as "legend(ring(0) col(1) pos(11) label(1 `e(G)'[1,1]))". The resulting graph seems magical to me, because it demonstrate "Total (Gini=0.286) Nonunion (Gini=0.294) Union (Gini=0.246)" all at once, which I think I should designate by typing such as "label(1 "Total"`e(G)'[1,1]) label(2 "Nonunion"`e(G)'[2,1]) label(3 "Union"`e(G)'[3,1])". So, can anyone explain what's the principle behind this? Thank you very much.

Code:
. sysuse nlsw88
(NLSW, 1988 extract)

. lorenz estimate wage, over(union) total gini

L(p)                              Number of obs   =      1,878

            0: union = nonunion
            1: union = union

--------------------------------------------------------------
        wage |      Coef.   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
0            |
           0 |          0  (omitted)
           5 |   .0162309   .0005027      .0152449    .0172168
          10 |   .0368529   .0008292      .0352268    .0384791
          15 |   .0601058   .0011475      .0578554    .0623562
          20 |   .0863558    .001589      .0832394    .0894722
          25 |    .114934   .0018728       .111261     .118607
          30 |   .1457833   .0022658      .1413395     .150227
          35 |   .1796046   .0026487      .1744099    .1847994
          40 |   .2155412   .0029961       .209665    .2214173
          45 |   .2541379    .003311      .2476444    .2606315
          50 |   .2953025   .0036368      .2881699    .3024351
          55 |   .3392766   .0039191      .3315903    .3469629
          60 |   .3866722   .0041917      .3784513    .3948932
          65 |   .4377062   .0044134      .4290505    .4463618
          70 |   .4930304   .0046064      .4839963    .5020645
          75 |    .553573   .0047252      .5443058    .5628402
          80 |    .620733   .0047993      .6113205    .6301455
          85 |   .6939954   .0047797      .6846212    .7033696
          90 |   .7744107   .0045737      .7654407    .7833808
          95 |   .8675309   .0036053      .8604601    .8746016
         100 |          1          .             .           .
-------------+------------------------------------------------
1            |
           0 |          0  (omitted)
           5 |   .0182581   .0007901      .0167085    .0198078
          10 |   .0411449   .0014039      .0383914    .0438983
          15 |   .0672693   .0020022      .0633426    .0711961
          20 |   .0959286   .0026353      .0907601    .1010971
          25 |   .1278324    .003394       .121176    .1344888
          30 |   .1627902   .0040507      .1548458    .1707346
          35 |   .2003534   .0046264        .19128    .2094268
          40 |   .2402022   .0051768      .2300492    .2503552
          45 |   .2826347   .0056621      .2715301    .2937394
          50 |   .3279457   .0061731      .3158389    .3400525
          55 |   .3757295   .0066285      .3627295    .3887296
          60 |   .4269463   .0070685      .4130835    .4408092
          65 |   .4811007   .0074622      .4664657    .4957358
          70 |   .5380573   .0077973      .5227649    .5533496
          75 |    .597479   .0081124      .5815688    .6133893
          80 |   .6604931   .0082738      .6442663    .6767199
          85 |   .7274785    .008437      .7109317    .7440253
          90 |   .7992943   .0082971      .7830218    .8155668
          95 |   .8809539   .0075437       .866159    .8957488
         100 |          1          .             .           .
-------------+------------------------------------------------
total        |
           0 |          0  (omitted)
           5 |   .0161147   .0004051      .0153202    .0169092
          10 |   .0367226   .0006934      .0353627    .0380826
          15 |   .0602741   .0010252      .0582634    .0622848
          20 |   .0868085   .0013371      .0841861    .0894308
          25 |   .1156008   .0016459      .1123728    .1188288
          30 |   .1472444   .0019984      .1433251    .1511636
          35 |   .1811803   .0022573      .1767531    .1856074
          40 |   .2178909   .0025612      .2128679     .222914
          45 |   .2572438   .0028595      .2516357    .2628519
          50 |    .299339    .003116      .2932277    .3054503
          55 |    .344591   .0033622       .337997     .351185
          60 |   .3931116   .0035971      .3860569    .4001662
          65 |   .4455101   .0037931       .438071    .4529493
          70 |   .5022412   .0039591      .4944765    .5100058
          75 |   .5640519   .0040864      .5560375    .5720663
          80 |   .6307203   .0041701      .6225418    .6388987
          85 |   .7026961    .004185      .6944883    .7109039
          90 |   .7810699   .0040762      .7730756    .7890643
          95 |   .8713064   .0033446      .8647468     .877866
         100 |          1          .             .           .
--------------------------------------------------------------

-------------------------
             |      Gini
-------------+-----------
           0 |  .2943305
           1 |  .2462299
       total |  .2861153
-------------------------

. matrix list e(G)

e(G)[3,1]
            Gini
    0  .29433046
    1  .24622995
total  .28611525

. lorenz graph, keep(total 0 1) overlay noci legend(ring(0) col(1) pos(11) label(1 `e(G)'[1,1])) o1(lpattern(solid) lcolor(red%50)) o2(lpatter(longdash) lcolor(blue)) o3(lpattern(shortdash) lcolor(green)) yaxis(1 2)
Array