If you know that your base date is the last date of the previous month, just add 1: date = input_date + 1; Otherwise, you could try: date = intnx('MONTH',input_date,1,'BEGINNING'); /Linus ... View more

See http://support.sas.com/kb/20/811.html. If you want to do this you have to write your own code. /Linus ... View more

Are you using SQL pass-thru? If not, I believe that you could use a SAS date literal, which SAS/ACCESS engine will convert automatically to a Teradata data value: Select * From some_table where mydate = &MVC_END_DATE; If using pass-thru, I think the easiest way is to create your macro variable in a data step: %let MCV_END_DATE='23Sep08'd; data _null_; call symput('MCV_REQUEST_END',"'"||put(&MCV_END_DATE,yymmdd10.)||"'"); run; Regards, Linus ... View more

(drop=h1:) /Linus ... View more

I believe that this has already been answered in a previous thread of yours...? /Linus ... View more

Who is you SAS representative (which receives SW and licenses)? For details help how to manage this contact SAS support. /Linus ... View more

This can be accomplished by using to sort steps. The first order you data by id and date (descending). The second will remove duplicate rows by id, which will keep the first row per id (with highest date value). /Linus ... View more

Could you please be more specific how your data is stored, and how you wish your result set would be like. PROC CONTENTS and some example data would help to understand your issue. /some ... View more

Well, you have to merge/join the table somehow. Do you want to match them based on observation number...? Some kind of Id column generally is the best/common way to do it. About the multiplying part: you have to convert your character column to numeric first by using the input function: input(sal,best.) /Linus ... View more

This should work if you move them to another directory, and if you don't rename them. Caution: this won't probably work if you chose to use SPDE in the future. /Linus ... View more

I'm not sure about EG 3, but in EG 4 you could use the Filter & Query object for this. I suggest that you'll take EG class, also available in the e-learning program. http://www.sas.com/apps/elearning/elearning_courses.jsp?cat=SAS+Enterprise+Guide Regards, Linus ... View more

I believe the following note will apply to this. The key is to add -allowxcmd to your object spawner startup (you are using a workspace server from EG, right?) http://support.sas.com/kb/14/696.html Regards, Linus ... View more

Just remember thai if you data is indexed on your "spid" is indexed or your data reside in an external RDBMS, you don't want any function calls in you where-clause. Use Scott's suggestion with %let, %sysfunc and putn function prior to your query (like in your initial example). Regards, Linus ... View more

I suggest that you build a macro that first read out all tables (and their column names) for a specific libname, using DICTIONARY.TABLES (from PROC SQL), then maybe do PROC FREQ on them to find out which columns has nonmissing values. Regards, Linus ... View more

... %let spids=1 and 30; .. where spid1 between &SPIDS.; ... /Linus ... View more