How did you create the dataset? Why did you make the variables as character if you wanted them to be numeric? ... View more

Hi @Ronein  You can use the scan function, which is the easiest way to do this job in your case. It splits your string in several parts ("words") according to a delimiter that you specify as the third argument (in your case, an hyphen) You can then retrieve the first, the second, ... word.    data tbl2; set tbl; Customer = scan(Customer_Branch,1,"-"); Branch = scan(Customer_Branch,2,"-"); run; ... View more

proc sql; create table work.macro_eco_new as select date, x, y, z If date >= "2002Q1",then select min(x) as x, min(y) as y from work.macro_eco ; quit; This is what i tried ... View more

OK. That would be more simple. Check if data is similar with yours.   data have; input id $ a b c d; cards; aa . . . . a1 1 . . . b2 2 1 . . c3 3 2 1 . ; proc iml; use have ; read all var _num_ into x[c=vname r=id]; close; x=choose(x=.,0,x); want=x+t(x); create want from want[c=vname r=id]; append from want[r=id]; close; quit; ... View more

MERGE has a specific meaning in SAS and I suspect you are conceptually using the word merge in a different meaning. Here is an example of a MERGE of two data sets in SAS with no common variables so you can see the result: data work.one; input x y; datalines; 1 2 3 4 6 6 ; data work.two; input w z; datalines; 33 44 55 66 ; /* no common variables*/ data work.merge1; merge work.one work.two; run; MERGE in SAS means to align rows side by side. If there are common values then the values from one data set will appear data work.three; input x z; datalines; 99 88 88 99 ; /* common variable*/ data work.merge2; merge work.one work.three; run; Notice that the values of X in the work.merge2 data set for the first two rows are the X values from work.three. Your merge meant that since you have the same variables in both sets that most likely you only see the "google" data.   What you want apparently is to STACK records from one set above the other. That is done in a data step with the SET statement. This shows what SET will do with data with no common variables and once common variable. data work.set1; set work.one work.two; run; /* or with one common variable*/ data work.set2; set work.one work.three; run; I suspect you're going to have some issues about lengths of common variables though. ... View more