08-31-2016 04:50 AM
I want to calculate No and percentage of firms for each country. (each firm is repeated by many time due to its data for several years)
My data contain several countries from year 2000 to 2015.
there is dummy variable of fam which is 0 for non family and 1 for family.
i want to make table in this format. These are just hypothetical values for reference. the attached file contain excel data in excel and cvs format. here percentage means percentage of family firms and non-family firms to overall sample firms.
Country Fam Non family
HKG 50 80
IDN 70 110
08-31-2016 05:02 AM
Please post your example data as data step.
08-31-2016 06:04 AM
Follow the instructions in this post to get a datastep text you can copy and paste into the text box:
08-31-2016 06:33 AM - edited 08-31-2016 06:34 AM
08-31-2016 06:11 AM
Post your data at this forum. No one would like to download a file from website. proc import datafile='/folders/myfolders/new.csv' out=have dbms=csv replace; run; proc format; value fmt 0-1=[percent8.2]; value fam 0='Non_Fam' 1='Fam'; run; proc sql; create table temp as select 1 as group,country,fam,count(distinct company) as v from have where country is not missing group by country,fam union all select 2 as group,country,fam, count(distinct company)/(select count(distinct company) from have) as v from have where country is not missing group by country,fam order by 2,1 ; quit; proc transpose data=temp out=want(drop=_: group); by country group; var v; id fam; format fam fam.; run; data want; set want; format _numeric_ fmt.; run;