The reason this post (Tom on 3/1/2013 at 2:41pm) finally cleared it up for me is that I realized that the functions MONTH, DAY, and YEAR were returning numeric values, which would automatically lack any leading zeroes, regardless of the formatting applied to the date they were extracting from. ... View more

Hi:   But that doesn't show what is in Status1. What does the data in Status1 look like? How is it created -- do you use a procedure to create STATUS1? It is in the WORK library. How does it get there? It looks like your DATA step program is only being used to keep the 4 variables. You didn't answer my question about what is row for "NOTH" in your example? Since you need both comma formats and percent signs in the Performance column, that is an argument for PROC  REPORT for the final report. But there are still too many unknowns for the report code, when the data manipulation is still not resolved. cynthia. ... View more

yes you may use the proc sql code above, it it requires a modification like below proc sql; create table pivot1 as select ln_no,category,count(ln_no) as type from summary1  group by category,ln_no; quit; you cannot use type for grouping, as you are generating the count in it. in your code it is not a category. Thanks, Jagadish ... View more

Hi Omega1983 I assumed that your variable name is var1 (if it is region then replace the code by actual varname). Thanks data col1; set col; var1=compbl(tranwrd(var1,' ','_')); var1=compbl(tranwrd(var1,'/','_')); count=1; run; proc sort data=col1; by var1; run; proc summary data=col1 missing nway; by var1; var count; output out=want(drop=_:) sum=; run; ... View more

If you want monthly data try yearmon7.  or yymm6. format for your dates which will group all data into months instead of days. ... View more

You don't need to import it, just use it. ... View more

Note that the comparisons in the where clause are case sensitive. proc sql; select var_date into :var_dates separated by  " " from sashelp.vtable where libname="SASHELP" and memname="STOCKS" and format="DATE9." ; quit; data want; set have; format &var_dates mmddyy10.; run; ... View more

If you are using the import task in Enterprise Guide there is a checkbox within the import process where you can specify to convert variable names to SAS variable names (or something like that). If you check this box then any special character or space is converted to an underscore for the variable name. This essentially sets the option that mentioned. ... View more

Hi:   This is probably a question for Tech Support. I do not believe there is a way to change the tab color with SAS, ODS and Tagsets.ExcelXP. I believe you do tab color changes in Excel with a popup window. SAS, generally, has no way to change popup windows. Another example of this limitation is adding table level style to the sheet -- in Excel, this is done using the pull-down menus inside Excel. SAS does have a way to change table style using ODS style templates, but NOT the same as using the Excel pull-down menus. cynthia ... View more

ods tagset should be odd tagset.excelxp perhaps? Usually means there's an error in the code somewhere but hard to see without all the code. If you're splitting the data by a variable, in a proc report you can do it in one step using the sheet_interval option and adding in a variable that has the name. ... View more

The SAS Code Analyzer might give something you can use? Base SAS(R) 9.4 Procedures Guide ... View more

If you truly have variables with spaces in the name then yes, the ''n notation should allow them to work, however SAS will sometimes convert them itself, but leaves the labels with the spaces. You can run a proc contents on your dataset to see if that happened. You can also try the option validvarname=any; to see if that helps.  ... View more

The environment version and usage is not mentioned. When using BI/DI server (EGuide) libnames can have very long names as defined  in the SAS metdatabase. The real used name is translated to the one having the 8 character limit. When the table is a physical location on the OS you can get its physical name. When it is RDBMS that is impossible. The statement:" Libname _all_ list ; "  will show all technical libname defiitions  _all_ is the * mask like indication it could be your specific libname. Showing .requesting details in EG will also let you view the details. The %sysfunc(pathname(yourlib)) will give the physical path of the libname on the Unix Server. The filelist option in EG will serve you with a lot of the same functionality as in Unix. http://www.bi-notes.com/2013/03/sas-enterprise-guide-unix-putty/ What is the thing you are trying to achieve? ... View more

Is this what you are looking to achieve? DATA SINGLEOBS;   SET LVG1;   BY LN_NO;   RETAIN _:;   ARRAY VARS{*} CPBFOR SNTLVG UPBFOR RG901 RG902 RG903 CK007 CK009;   ARRAY _VARS{*} _CPBFOR _SNTLVG _UPBFOR _RG901 _RG902 _RG903 _CK007 _CK009;   DO  I = 1 TO DIM(VARS);   IF FIRST.LN_NO THEN _VARS{I} = .;   IF VARS{I} THEN _VARS{I} = VARS{I};   END;   IF LAST.LN_NO THEN OUTPUT;   DROP _: I COMM_CD LTR_ID ;   FORMAT CPBFOR SNTLVG UPBFOR RG901 RG902 RG903 CK007 CK009 DATE9.; RUN; ... View more

data test(rename=('A Category'n=A_Category 'B Category'n=B_Category)); Or better yet, since you are not performing any tasks in the small data step code other than renaming, use PROC DATASETS to rename your variables. ... View more