This has an error   top_tx_min_cible_&var = tx_min_cible_&var.sum;   It should say   top_tx_min_cible_&var = tx_min_cible_&var..sum;   Note there must be two dots after &var on the right hand side of the equal sign, one dot to end the name of the macro variable, and one dot to indicate the dot that PROC REPORT requires in this situation before the word SUM. You need to fix this throughout your code. From now on when you get an error in code, you MUST (it is not optional) show us the entire log for this PROC, every single line in the log for this PROC, not just the error messages. If the above still doesn't fix your problem, we need to see the ENTIRE log for this PROC, not just the error messages. In addition, you should turn on macro debugging tools by running this line of code before you run the macro:   options mprint; ... View more

Divide by 1000?   Picture format? ... View more

This solution assumes that the values of startdate and enddate are valid numeric SAS date values.   data have; startdate='03JAN2022'd; enddate='05MAY2022'd; run; data want; set have; diff_months=intck('month',startdate,enddate); do i=0 to diff_months; if i=0 then newcolumn=startdate; else if i=diff_months then newcolumn=enddate; else newcolumn=intnx('month',startdate,i,'e'); output; end; drop i diff_months; format newcolumn date9.; run; ... View more

@kcvaldez98 wrote:   ERROR: There is more than one statistic associated with the column defined by the following elements. You can only have one statistic per variable in a DEFINE statement. You can't request two (or more) statistics per variable in a single DEFINE statement.   In PROC REPORT, you can use an ALIAS to get multiple statistics for a single original variable. See this example:  https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/proc/n0ldy84v9j7463n1jjkloe7av38c.htm ... View more

Creating variables named JAN1, JAN2, JAN3, JAN4 is a particularly poor way of arranging the data in SAS, and makes future programming with this data set more difficult and time consuming. Dates are data, they belong as values of a variable, not in a variable name.   A much better approach is to leave the data in the long arrangement. This assumes moveindate and moveoutdate are valid numeric SAS date values.   data re_arrange; set have; do date=movindate to moveoutdate by 1; output; end; drop moveindate moveoutdate; run;   Further calculations with this data set are much easier because almost all SAS PROCs work well on data in this long arrangement.   Please answer this question: the next steps after this re-arrangement of the data will be to perform analyses or reports or plots like _______________ and ________________. ... View more