should be written in format RP /AA/BB/CC, for should be written in format PR/ AA/BB/CC and report No. shall be written in right side corner in front of validation protocol/validation report with font type Times New Roman with font size 11.
should be written in format RP/VL/AA/BB/CC, be written in format PR/VL/AA/BB/CC and report No.
Hope this helps in proceeding some code which should return the result you're after. Snapshot_dte = mdy(mod(snapshot_date,100),1,int(snapshot_date/100)) ĭata credit_data (keep=snapshot_dte acct_id BQR BQR_1-BQR_12 default_flag write_off_flag) That means, the SAS code should be such that internally it loop for next 12 months per row and create default flag This process is to be executed for each row forward looking into 12 months data and creating default_flag Similarly, when row2 - acct_id = 123 BQR=5 and Snapshot_dte = 01FEB1990 is read, the default flag is created in such a way that in next 12 month, that is SAS would read all records related to this acct_id from 01FEB1990 to 01JAN1991 and check if BQR=9 ever in this 12 month time and flag the variable default_flag = 1, if found else 0 if not found against row2
This default flag is created in such a way that in next 12 month, that is SAS would read all records related to this acct_id from 01JAN1990 to 01DEC1990 and check if BQR=9 ever in this 12 month time and flag the variable default_flag = 1, if found else 0 if not found against row1 I want to create default_flag for row1 - acct_id = 123 BQR=4 and Snapshot_dte = 01JAN1990
I will take up remaining 2 queries in next thread one-by-one.
Thanks for the reply, but this is not what I wanted. Should you be dealing with large data volumes then tweaking for performance would be possible for the posted code - but it would complicate the syntax so it's only something I'd be doing if really necessary. If write_off_flag='0' and default_flg='1' thenĭo _i=1 to &write_of_months while (_rc = 0) * populate bqr1 to bqr12 and default_flg */ Retain default_flg '0' write_off_flag '0' ĭcl hash h1(dataset:cats('have (where=(acct_id="',acct_id,'"))'),ordered:'y') That's certainly later than 60 months ahead of the date where BQR gets a value of 'W'. Looking into your Excel I don't understand why the write_off_flg gets only so late set to '1'. Some code which should return the result you're after.