Kemal Aslan wrote:
For maxage and bromaxage, my main motivation was to take into
account twin, triple kids. But, unfortunately, I can't use then
efficiently in the first place to derive the number of older boys.
... could you give me any advice how I can take into acoount twin,
triple kids?
---------------------------------------------------------
OK, this took a couple of trials, but I think I ended up right.
Check carefully!
Best wishes,
Svend
clear
input str4 hhid hrelhd age female
10_7 1 43 0
10_7 2 38 1
10_7 3 19 0
10_7 3 16 1
10_7 3 13 1
10_7 3 11 0
10_8 1 43 0
10_8 2 38 1
10_8 3 10 1
10_8 3 11 0
10_8 3 11 0
10_8 3 13 0
10_8 3 14 1
10_8 3 14 0
10_8 3 14 0
10_8 3 14 0
end
gen male = 1-female
// by...: does not consider age sorted after gsort -age.
gen _age = -age
sort hhid hrelhd _age
// -boysum- is boy number (oldest first); twins have different numbers
by hhid hrelhd: gen boysum = sum(male) if hrelhd==3
// -boysum2- is boy number (oldest first), but twins have the same high number
by hhid hrelhd _age: egen boysum2 = max(boysum) if hrelhd==3
// -boysamage- it the number of boys of same age
by hhid hrelhd _age: egen boysamage = sum(male) if hrelhd==3
// The number of older brothers is boysum2-boysamage
gen nolderbr = boysum2-boysamage
sort hhid hrelhd age
list hhid-female boysum-nolderbr
+-----------------------------------------------------------------------+
| hhid hrelhd age female boysum boysum2 boysam~e nolderbr |
|-----------------------------------------------------------------------|
1. | 10_7 1 43 0 . . . . |
2. | 10_7 2 38 1 . . . . |
3. | 10_7 3 11 0 2 2 1 1 |
4. | 10_7 3 13 1 1 1 0 1 |
5. | 10_7 3 16 1 1 1 0 1 |
|-----------------------------------------------------------------------|
6. | 10_7 3 19 0 1 1 1 0 |
7. | 10_8 1 43 0 . . . . |
8. | 10_8 2 38 1 . . . . |
9. | 10_8 3 10 1 6 6 0 6 |
10. | 10_8 3 11 0 5 6 2 4 |
|-----------------------------------------------------------------------|
11. | 10_8 3 11 0 6 6 2 4 |
12. | 10_8 3 13 0 4 4 1 3 |
13. | 10_8 3 14 0 1 3 3 0 |
14. | 10_8 3 14 0 3 3 3 0 |
15. | 10_8 3 14 0 2 3 3 0 |
|-----------------------------------------------------------------------|
16. | 10_8 3 14 1 0 3 3 0 |
+-----------------------------------------------------------------------+
________________________________________________________
Svend Juul
Institut for Folkesundhed, Afdeling for Epidemiologi
(Institute of Public Health, Department of Epidemiology)
Vennelyst Boulevard 6
DK-8000 Aarhus C, Denmark
Phone, work: +45 8942 6090
Phone, home: +45 8693 7796
Fax: +45 8613 1580
E-mail: [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/
