You could call this combining values by ID. I'm sure there's a better name.   data have ; input ID VALUE$; datalines; 101 V1 101 V2 101 V3 102 V4 102 V5 ; run; data want (rename = (values = value) drop = value); do until (last.id); set have; by id; put value = ; put id = ; if first.id then values = value; else if not first.id then do; values = catx(" ", values, value); put values = ; put id = ; end; end; run; ID value 101 V1 V2 V3 102 V4 V5 Note that your data set will need to be sorted by ID.   I used the PUT statements so that you can look at the log to see how it is working internally.   69 data want (rename = (values = value) drop = value); 70 do until (last.id); 71 set have; 72 by id; 73 put value = ; 74 put id = ; 75 if first.id then values = value; 76 else if not first.id then do; 77 values = catx(" ", values, value); 78 put values = ; 79 put id = ; 80 end; 81 end; 82 run; VALUE=V1 ID=101 VALUE=V2 ID=101 values=V1 V2 ID=101 VALUE=V3 ID=101 values=V1 V2 V3 ID=101 VALUE=V4 ID=102 VALUE=V5 ID=102 values=V4 V5 ID=102 ... View more

You can't use LAST.Breakhigh because it isn't established yet in the code. In other words, it does not exist in your CSRP data set. The variable exists in HIGHS, so you would need to use FIRST/LAST on the next data set you create. This comes down to how the PDV works.    FIRST/LAST processing can occur when the variable exists on the input data set, you have sorted it by the desired variables, and you include it on the BY statement after the SET statement.   Also, please provide usable data for us. You can use the data to data step macro. ... View more