Tested this, PROC GLM does include all data when available, where PROC REG excludes any row with missing data.  So PROC GLM will avoid that issue entirely, which is good to know! Thanks @PGStats   title 'Two at once'; proc glm data=class; model weight height = age; run; title 'Weight regression'; ods output parameterestimates; proc glm data=class; model weight = age; run; title 'Height regression'; ods output parameterestimates; proc glm data=class; model height = age; run; ... View more

I guess you may need to do:   proc sql; create table master_table as select a.*, b.r_size, b.size_min, b.size_max, b.date as date_b from dataseta a left join datasetb b on a.permno=b.permno and year(a.date) = year(b.date); quit; ... View more

Similar to the suggestion by @Cynthia_sas (as far as I know, in SAS Studio WRDS one cannot directly export to the local machine, unless one is connected to the WRDS server with his or her local machine using SAS rather than SAS Studio and uses a program like WinSCP), you can do the following trick: libname out "/home/.../"; data out.filename; set filename; run; Or you can do the following: data '/home/.../filename.sas7bdat'; set filename; run; Both should work. This exporting keeps your data in the .sas7dbat format. However, you can also use the following command if you want to export your data in .csv format: proc export data=work.filename DBMS=csv outfile="/home/.../filename.csv" replace; run; I hope this helps @fafrin420.   ... View more

Hi, Good afternoon. I am working on the same replication paper: SP Kothari 2005. Could you tell me if the you have encountered the problem that a performance matched firm can bot be obtained? I mean I used your code with the modification of Ksharp. I did not find any missing var for the ***PDAs. ... View more