******************************************************************
******** This program generates the G2 population ****************
******************************************************************

clear

forva x=2006/2017{
use E:\Data\rawdata\705830\bef`x'
keep pnr foed_dag`x' alder`x' ie_type`x' koen`x' /*keeps identifier, birth year, age, immigration status and gender */
keep if alder`x'>21 /*keep only individuals older than 21 */
rename alder`x' alder
rename ie_type`x' ie_type 
rename koen`x' koen
gen fodaar=year(foed_dag`x')
save E:\Data\workdata\705830\zic\sha\replication\data\bef`x', replace
}

use E:\Data\workdata\705830\zic\sha\replication\data\bef2006

forva x=2007/2017{
append using E:\Data\workdata\705830\zic\sha\replication\data\bef`x' 
}

save E:\Data\workdata\705830\zic\sha\replication\data\bef_samlet_test, replace

clear

use E:\Data\workdata\705830\zic\sha\replication\data\bef_samlet_test

bysort pnr: gen optal=_n

bysort pnr: egen max_optal=max(optal)

keep if max_optal==optal /*keep latest observation per individual - to avoid duplicates and count the number of data years per individual */

replace fodaar=0 if fodaar==.
replace ie_type=0 if ie_type==.
replace koen=0 if koen==.

bysort pnr: egen fod_max=max(fodaar)
bysort pnr: egen ie_max=max(ie_type)
bysort pnr: egen koen_max=max(koen)

drop fodaar ie_type koen 

rename fod_max fodaar
rename ie_max ie_type
rename koen_max koen

keep pnr fodaar ie_type koen max_optal /*keep identifier, birth year, immigrant status, gender and data years */

save E:\Data\workdata\705830\zic\sha\replication\data\grund_pop, replace