Stata Technical Bulletin
Options
gp(gpname) creates a new categorical variable, gpname, containing categories summarizing group membership.
Example
To produce output mimicking the UK official low income statistics, we use the mean income as the cut-off value input into
xfrac:
. summarize eybhc [fw=wgt]
Variable ∣ Obs Mean Std. Dev. Min Max
---------+-----------------------------------------------------
eybhc I 5.6e÷07 233.0179 199.0178 -123.9898 7740.044
. local mean = .result(3)
. xfrac eybhc [fw=wgt], cut('meanz) gp(fracgp)
Warning: eybhc has 20 values < 0. Used in calculations
Proportions of the sample in subgroups defined
by values of eybhc between specified fractions
of a cut-off value = 233.01790
Fractions of∣ |
Freq. |
Percent |
Cum. |
— | |||
<.1 I |
4ББ1Б2 |
0.81 |
0.81 |
.1-.2 I |
482238 |
0.86 |
1.68 |
.2-.3 I |
912Б26 |
1.63 |
3.31 |
.3-.4 I |
3983433 |
7.13 |
10.44 |
.4-.5 I |
ББ02971 |
9.8Б |
20.30 |
.5-.6 I |
Б186Б97 |
9.29 |
29. Б8 |
.6-.7 I |
493ББ14 |
8.84 |
38.42 |
.7-.8 I |
4777040 |
8.ББ |
46.97 |
.8-.9 I |
4341904 |
7.77 |
Б4.7Б |
.9-1.0 I |
4364218 |
7.81 |
62. Б6 |
1.0-1.1 I |
3234833 |
Б.79 |
68. ЗБ |
1.1-1.2 I |
2678779 |
4.80 |
73.1Б |
1.2-1.3 I |
26БББ24 |
4.7Б |
77.90 |
1.3-1.4 I |
209Б389 |
3.7Б |
81.66 |
1.4-1.5 I |
1683166 |
3.01 |
84.67 |
1.5-1.75 I |
3149798 |
Б.64 |
90.31 |
1.75-2.0 I |
1848821 |
3.31 |
93.62 |
2.0-2.5 I |
19020Б9 |
3.41 |
97.02 |
2.5-3.0 I |
721933 |
1.29 |
98.32 |
>=3.0 I |
939810 |
1.68 |
100.00 |
—— — ——— — ——— —-⅛-- |
— | ||
Total I |
ББ8Б170Б |
100.00 |
There is no official poverty line in Britain, but half of the average income is used by many commentators as such a threshold.
The xfrac output shows that about one fifth of the UK population in 1991 had incomes below one half of contemporary mean
income (and 62.6% had incomes below the mean). But observe too that 38% of the population have incomes between 40%
and 60% of mean income. Thus relatively small changes in the threshold defining the poverty line can have a large impact on
estimates of the proportion who are “poor”.
The command above also created a new variable summarizing income group membership. If we were now to type
. table fracgp tenure [fw=wgt], row col
we could compare the shape of the income distribution across housing tenure groups.
ineqdeco, ineqdec0: inequality indices, with decompositions by population subgroup
ineqdeco and ineqdecθ estimate a range of inequality and related indices commonly used by economists, plus decom-
positions of a subset of these indices by population subgroup into within- and between-group inequality components. Inequality
decompositions by subgroup are useful for providing inequality profiles at a point in time, and for analyzing secular trends using
shift-share analysis. Unit record (micro level) data are required. For a non-technical introduction to the topic, see Jenkins (1991).
Standard textbook treatments are provided by Cowell (1995) and Lambert (1993).
Inequality indices estimated by ineqdeco are: members of the single parameter Generalized Entropy class GE(α) for
a = -1,0,1,2; the Atkinson class A(e) for e = 0.5,1,2; the Gini coefficient, and percentile ratios such as p90∕p10 and
p75∕p25. Also presented are related summary statistics such as subgroup means and population shares. Optionally presented are