dates

[unnati]

Hello,

I am trying to find maximum value from all dates variable which is present in Macro variable"  &dates. ". Can someone help me how to find Maximum value from this all dates. I convert all dates in date9. format by using following code.  I wrote following code to find maximum value but it gives me error message. 

Any help will be appreciated .

Thank you in Advance.

****************************  DATA SET STATEMENT ***********************

data now;
set nn;
format &dates date9.;
informat &dates date9.;
where Subject^='';
run;
proc sort data=now out=new_; by subject; run;

data ll;
set now_;
date1=max(&dates);
format date1 date9.;
run;

*******************  LOG ERROR MESSAGE *********************************

ERROR: File WORK.NOW_.DATA does not exist.
35 date1=max(&dates);
NOTE: Line generated by the macro variable "DATES".
35 EGDT EGDT_AT6 EXEGDT LSTSMKDT ACCUDT ASESDT CNTRDT EDSSDT EVALDT EXAMDT HPTEVLDT PFDT PFMKDT PFTDT SDMTASDT DSDT PGNDT
________
388
76
35 ! LBDT PEDT PEASSDT ESSDT HADSDT PWCDT SUICDT QUEDT VSDT VSNDT MRIDT AESTDT AEENDT VISDT CMSTDT1 CMENDT1 ENSYMDT MSRLDT
35 ! RELAPDT STSYMDT
ERROR 388-185: Expecting an arithmetic operator.

ERROR 76-322: Syntax error, statement will be ignored.

[ballardw · Accepted solution]

First thing is fix errors:

proc sort data=now out=new_; by subject; run;

data ll;
set now_;
date1=max(&dates);
format date1 date9.;
run;

The second data step did not reference the sorted data.

Second the data step MAX function wants variable names to be comma delimited:

y = max(a, b, c);

If you do not know how many variables you are using then likely an array would be easier

data ll;
   set new_;
   array d__ &dates.;
   date1=max(of d__(*));
   format date1 date9.;
run;

I used a name like d__ for the array as I have no idea what other variables you may have and didn't want a complaint about an error generated because the array name I picked matched an existing variable. If you have a variable named d__ then change the references above to something NOT in your data.