I think the issue was because of confusion between how SAS internally stores the data and how it shows it to user (both as Output Data and Exported csv/xlsx data). I removed all the Formats and Informats from the data1 import process after editing the copy of the import task source code. Removed all the `FORMAT 20.2` references from data2 proc sql code and combined both of the datasets and then rounded off all the columns having decimal values. After that proc sort nodupkeys by = _all_ method worked flawlessly and removed all the duplicates.
... View more
