That is a part of SAS BI Server package. You need to install SAS components with your MS Office client installation. It means that you can use Excel as a client to connect to SAS via Intelligent Architecture Servers. You can access data directly or running Stored Processes, among other thing. See www.sas.com and support.sas.com for more extensive information about this topic. /Linus ... View more

By doing so, I think you just created a HTML-file with a XLS suffix, but it's still a HTML-file. You need to open this file into Excel, and then Save as.. Microsoft Excel Workbook. But why don't you just use the OUT= statement in PROC MEANS? It seems quite unnecessary complicated... /Linus ... View more

I suggest that you share your PROC IMPORT code together with the SAS LOG, and maybe a description about how your HTML data is placed in the Excel file. /Linus ... View more

Do you mean that you have a HTML-file that you have imported to Excel, and then save as an Excel file? In this case you could good old PROC IMPORT or LIBNAME EXCEL. /Linus ... View more

The CALL EXECUTE will be executed after the calling data step has finished, so in your case the assignment will be performed before the SQL that will calculate the value for the macro variable. This is because a SQL cannot run within a data step... /Linus ... View more

> What are the career prospects for SAS Enterprise > Miner? Contact some consultancy's that is working with analytics. They should be able to give the picture of this "marketplace". > > what are the best features of enterprise miner > that makes it competitive with other software > available in market? There is a lot of papers/articles published on this matter. Just search via Google or the search engine of tour choice. > > How is Enterprise Miner different from SAS > forecasting studio or SAS ETS? On www.sas.com there is lot of material available which describe the different SAS products. /Linus ... View more

I suggest that you contact SAS technical support for this kind of issues. Regards, Linus ... View more

A search on Google on "nonparametric test sas" gave this among others: http://analytics.ncsu.edu/sesug/2004/TU04-Pappas.pdf. Sane search on support.sas.com gave 409 answers, I'll guess you might find some information there that will help you out. /Linus ... View more

The eq operator can only be used in comparisons, in IF and WHERE for example. The = operator can be used both in comparisons and in assignments. Some people think it's easier to read the code if using different operators in assignment and comparison statements, but technically, there is no difference. /Linus ... View more

Try this: data visits; input id date; informat date mmddyy10.; format date mmddyy10.; datalines; 001 10/01/2007 001 01/02/2008 002 01/15/2008 002 02/20/2008 002 08/09/2008 ; run; data vistsno; set visits; by id; if first.id then noVisit = 0; noVisit + 1; run; /Linus ... View more

Prior to the current release, you could do something similar by using a HOLAP architecture. With SAS9, OLAP was totally rebuilt. One result was much better performance, so I guess SAS didn't think that there was no need for this functionality any more. There is some new features in 9.2, but I don't this is not one of them. One feature of interest is that will be able to incrementally update a cube (but the result will still be on physical/logical cube). /Linus ... View more

Well, you can use in-line views in your set statement together with a where-clause. See my example on SASHELP.CLASS: data classGrade; input name $ grade $; datalines; Alfred A Barbara C Janet D ; run; data class; set sashelp.class; length eval $6; run; PROC SQL; update class as a set eval = (select case when grade = 'A' then 'GOOD' when 'D' then 'BAD' else 'MEDIUM' end from classGrade as b where a.name eq b.name) where a.name in(select name from classGrade) ; quit; If your data is large, I think performance could be an issue with this SQL. To prevent long running updates, index your transaction table on the id column(s). Another option is simply to recreate the table using a join to do your calculation. /Linus ... View more

Well, it depends on your programing environment (i.e. are you using Enterprise Guide, are you using the intelligence architecture with metadata and workspace servers). For starters you may do the Essentials course, and then may take individual lessons depending on your needs and specific tasks that you will perform. Regards, Linus ... View more

It seems that you cannot do that. So your choices are: - run multiple imports, and then append the different result sets - read the whole Excel-sheet and post process it - use the LIBNAME excel engine together with named ranges, each named range will appear as separate SAS tables, which you could read in one operation (data-set a b c...;). /Linus ... View more

You'll need to have the Oracle SQL*NET client installed on your SAS client machine. The path could be an DB alias that is specified in the TNSNAMES.ORA file, see this note: http://support.sas.com/kb/23/865.html Regards, Linus ... View more