So its a number, not a character date. Change the formulas as below in all the relevant places.
input(put(a.regis_dt, 8. -l), mmddyy10.)
Current code:
proc sql;
CREATE table ccases1a
as SELECT
a.study AS study_num, i.inst_id, a.patient AS patient_id, a.regis_dt, a.pt_sex AS sex_id, a.ethnicity AS ethnic_id
FROM calgb.rss_accrual as a
JOIN calgb.institution as i
ON a.inst_nci_id = i.nci_id
WHERE i.GROUP_ID = 1
and input(put(a.regis_dt, 8. -l), mmddyy10.) >= &start.
AND (input(put(i.stop_dt, 8. -l), mmddyy10.) =. or input(put(a.regis_dt, 8. -l), mmddyy10.) <= &end.);
quit;
current error:
ERROR: Numeric format F in PUT function requires a numeric argument.
ERROR: Numeric format F in PUT function requires a numeric argument.
NOTE: The SAS System stopped processing this step because of errors.
NOTE: PROCEDURE SQL used (Total process time):
real time 0:00.01
cpu time 0:00.01
this means that stop_dt and regis_dt don't have the same formats. Clarify this. If it is a number use the one above, if its a character use the first one.
They are character fields. Here is what we have now.
24595 proc sql;
24596 CREATE table ccases1a
24597 as SELECT
24598 a.study AS study_num, i.inst_id, a.patient AS patient_id, a.regis_dt, a.pt_sex AS sex_id,
24598! a.ethnicity AS ethnic_id
24599 FROM calgb.rss_accrual as a
24600 JOIN calgb.institution as i
24601 ON a.inst_nci_id = i.nci_id
24602 WHERE i.GROUP_ID = 1
24606 AND input(a.regis_dt, yymmdd10.) >= &start.
24607 AND (input(i.stop_dt, yymmdd10.) =. or input(a.regis_dt, yymmdd10.) <= &end.);
ERROR: INPUT function requires a character argument.
ERROR: Expression using equals (=) has components that are of different data types.
24608 quit;
Is that the code you are referring to for character strings?
The second error had to do with the "=.". I changed that to = '' " and we're ok then.
Interestingly enough, this code works (in a data block).
where &start. <= input(regis_dt, yymmdd10.) <= &end.;
May the fact that it is contained in the sql code that is causing the problem.
I got around it by doing this, which does not give me any errors:
proc sql;
CREATE table ccases1a
as SELECT
a.study AS study_num, i.stop_dt, i.inst_id, a.patient AS patient_id, a.regis_dt, a.pt_sex AS sex_id, a.ethnicity AS ethnic_id
FROM calgb.rss_accrual as a
JOIN calgb.institution as i
ON a.inst_nci_id = i.nci_id
WHERE i.GROUP_ID = 1 ;
quit;
data ccases1a; set ccases1a;
where input(regis_dt, yymmdd10.) >= &start.
AND (stop_dt =. or input(regis_dt, yymmdd10.) <= &end.);
run;
So basically, stop_dt is a valid SAS date and regis_dt is a character field already.
Glad you got it working.
I have to check the data but stop_dt is missing in the output. I have to see if it existed in the input.
Thanks for all your help. It appears that pulling the test out of the SQL made it all work.
Your problem seems to be that you do not know what format you dataset variables are in. Use PROC CONTENTS and check. If they are numbers with a format like DATE, MMDDYY, YYMMDD, etc attached to them then they are DATES and you can compare them with a macro variable by setting the macro variable in DATE9 format and using it in a date literal.
%let start=01MAR2013 ;
.... a.regis_dt > "&start"d
If your macro variable START is in some other format then you can use the INPUT (or INPUTN) function to convert it to an actual date. There is no need to format it as you will be comparing the number and not the way the number is displayed.
%let start=2013-03-01 ;
... a.regis_dt > input("&start",yymmdd10.)
If your dataset variables are NOT dates then you can apply functions to them also to convert. For example if instead they are DATETIME values then use the DATEPART() function to convert them to date values. If they are character strings then use the INPUT function to convert them to dates.
%let start=01MAR2013 ;
.... input(a.regis_dt,yymmdd10.) > "&start"d
Don't miss out on SAS Innovate - Register now for the FREE Livestream!
Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.