Hello Statalisters
I created a pseudo panel by collapsing variables by birth_year and Gender for the 4 population groups in the dataset I am working with. The result is that the pseudo panel has equal proportions of the the 4 population groups, i.e each race makes up 25% of the total in the pseudo. I used data from 2 cross sectional surveys of which the original percentages after weights are known. How best can I weight my pseudo panel to replicate the percentages in the original datasets?
Below is part of the pseudo panel.
* Example generated by -dataex-. To install: ssc install dataex
clear
input float Race byte Gender float(birth_year Year)
1 2 1934 2014
1 2 1936 2008
1 2 1937 2014
1 1 1938 2008
1 1 1939 2008
1 2 1940 2014
1 1 1941 2008
1 2 1942 2008
1 1 1943 2014
1 1 1944 2008
end
label values Gender Q12SEX
label def Q12SEX 1 "Male", modify
label def Q12SEX 2 "Female", modify
[/CODE]
Thank you
I created a pseudo panel by collapsing variables by birth_year and Gender for the 4 population groups in the dataset I am working with. The result is that the pseudo panel has equal proportions of the the 4 population groups, i.e each race makes up 25% of the total in the pseudo. I used data from 2 cross sectional surveys of which the original percentages after weights are known. How best can I weight my pseudo panel to replicate the percentages in the original datasets?
Below is part of the pseudo panel.
* Example generated by -dataex-. To install: ssc install dataex
clear
input float Race byte Gender float(birth_year Year)
1 2 1934 2014
1 2 1936 2008
1 2 1937 2014
1 1 1938 2008
1 1 1939 2008
1 2 1940 2014
1 1 1941 2008
1 2 1942 2008
1 1 1943 2014
1 1 1944 2008
end
label values Gender Q12SEX
label def Q12SEX 1 "Male", modify
label def Q12SEX 2 "Female", modify
[/CODE]
Thank you
