> [email protected]
> >
> > Yesterday night, I encounter this problem. I have the
> > following table:
> >
> > . ta senum rich3
> >
> > Enumeratio | rich3
> > n Area | 1 2 3 | Total
> > -----------+---------------------------------+----------
> > 1 | 45 31 26 | 102
> > 2 | 13 10 6 | 29
> > 3 | 40 18 4 | 62
> > 5 | 11 13 20 | 44
> > 2171 | 0 10 1 | 11
> > 2455 | 0 0 11 | 11
> > 2920 | 0 11 0 | 11
> > 3052 | 0 1 10 | 11
> > 4163 | 0 9 3 | 12
> > -----------+---------------------------------+----------
> > Total | 2091 2085 931 | 5107
> >
>
> > I am interessed in the first column (very poor). I want to
> > compute the ratios
> > of poor in the psu
> > (for example for the first line: (45 /102 )*100 , for the
> > second line: (13 /29)
> > *100, etc...)
> > and group the PSUs into 3 categories: 40% of poor, 60% of
> > poor and more than
> > 80% of poor.
> >
> > I use the command:
> > tab senum rich3, matcell(mat_1)
> >
> > to retrive the matrix but I don't know how to retrive the vector.
> >
> > I just copy the table into Excel to compute my ratios.
> >
> > Is there any way to proceed in stata?
>
> If you proceed this way, you end up having
> to do an elementwise division of vectors;
> that's possible (-search matewd-) but awkward; and then
> you have to map your results back onto the dataset.
>
> Here is another way:
>
> egen vpoor = sum(100 * (rich3 == 1)/_N), by(senum)
>
> Watch out, however, for missings.
This may be more transparent. It
is also better if there are missings.
egen num = sum(rich3 == 1), by(senum)
egen den = sum(rich3 < .), by(senum)
gen vpoor = 100 * num / den
Nick
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
