Hi, It is a "union all". The result dataset will contain all observation of both datasets. ... View more

Hi Mark,   You could first of all, do a proc sort by colX and ColY. After that, you could create a new variable order=_N_.   Doing this you create a non missing variable to order data. adding to columns this variable and a definition like   define Order /noprint order   you should see your data as you want. ... View more

Hi,   I think that the problem is your macro variable path1. You could not introduce a variable content in a data step with a %let.   When you do that, you are not putting office or path contents into macro variable, you are literaly putting  &off is replaced by office and &path1 by path.   %MACRO CREA_XML ; *Create number of observations; proc sql noprint; select count(*) into :_nobs from dataset; quit; %local i; *For each observation:; %do i=1 %to &_nobs; *Keep path; data _null_; set dataset; if _N_=&i.; path = cat('"',"'",'/directory /folder/',strip(office),'.xml"'); call symput('path1',path); *this put the content of the last observation of path into macro variable path1; run; *And generate xml with this info; data _null_; SET dataset; file "&path1."; PUT '<?xml version="1.0" encoding="UTF-8"?>'; PUT ' '; PUT '<!DOCTYPE UPDATE SYSTEM "doc.dtd">'; PUT ' '; PUT '<DOC>'; PUT ' <JOB>'; line = ' <APP FROM="'|| trim(office) || '/>'; PUT line; PUT ' </JOB>'; PUT '</DOC>'; run; %end; %MEND CREA_XML ; Try something like this   ... View more

Hi CP2,   Seems a Machine-Precision probleme. It's related with how number are saved in base 2 in a compute, for example 0.1 could not be represented without error in a compute.    It's happen in all compute with all numbers, a way to solve it could be a round(var,0.0000000000001). Since you are working with not many decimals, this should not affect your calcs and solve the problems with EG.     ... View more

In macro language, %let is use to set the value of a variable into something. For example, %let i=1; will set the macro variable i to 1, so when you write &i. will be replaced by 1. %scan do the same that a scan function in a data step. In that case, select the &i word, separated by default separators. ... View more

Hi, I don't use VA but I hope that I can help with any idea.   Maybe if you replace the value 0 by 0.01 or something similar, this value will appear, and should be enought. ... View more

Hi, I don't use Visual analitics, but in SAS, to change the format you use something like put(id_project, best8.); Indicating that id_project must be put in best8. The result of the put is a string with the content of id_project in the format that you requested in categorical.     ... View more