SETINIT syntax or output? I didn't think that PROC SETINIT even had any options. The SETINIT output can be cryptic because it describes the "packages" that have been licensed, not the individual procedures or products. The expiration dates are clear. To find out the specific procedures available (or licensed, as they can be different) on a SAS server, you could search comp-soft-sys.sas (in Google groups); it's been discussed there multiple times. It may also be in a SAS note on support.sas.com. ... View more

Helen, Chuck's right; it's a Unix shell script. I'm not sure of which flavor of Unix you have, but I do know that Solaris 9 does not support data compression through the server (Solaris 10 does; it may be worth the migration for that one feature). When a file is gzip'd, there is typically an 80% savings in disk space on a SAS dataset, so there is significant value to the Unix admin to having the files as .gz (in particular, for backup times). It's a bit of a pain for EG users storing the data on Unix. A kluge work-around would be to have a personal Unix scheduled task (cron) to sweep through the data directories and do a daily gunzip of the files (gunzip changes the non-root owner of the files, so the user needs write and delete access to the directory). To find the script, the Unix admin's are probably going to have to go through the root user's crontab. Doc ... View more

You may need to be more specific in your question. I can have multiple code tasks open at once, but only one analysis task. ... View more

I'm not sure why you are trying to do this directly in code. If you are using EGuide (the topic of this group), the compare task is under the data tab. It leads you through it and sorts and sets up the keys for you. ... View more

The option you are looking for is called MPRINT. ... View more

Hippo, I doubt you'll save any time in in the second question (and better to start a new thread for a new question), by using a DATA step. You will need to sort both data sets (doing the K subsetting in the SORT), and then use a data step to merge them (as I showed code for earlier). Then you can use MEANS to output the statistics (if both variables are numeric) or TABULATE if they are a mix. Alternately, you could use RETAIN coding in the data step to aggregate the statistics. These are moderately advanced topics, so you may want to read up in either Aster & Seidman's "Professional SAS Programming secrets" or Jaffe's "Mastering the SAS System" (both were available from the SAS publications web site). Doc Muhbaier Duke ... View more

What you want is a MERGE, not a SET. * untested code; PROC SORT DATA=table1; BY field1; RUN; PROC SORT DATA=table2; BY field2; RUN; DATA test; MERGE table2 (in=int2) table1 (in=in1); BY field1; IF in1; RUN; You may get slightly different results if there are repeated values of field1 (particularly if there are matching repeats in both datasets). Doc Muhlbaier Duke ... View more

The contents of the tables are largely driven by the topic under consideration. Efficacy measures should be in the protocol. Safety measures always include death, but there is little else that is an 'always'. Items to consider in safety are also topic driven; the safety items for a statin might be quite different than those for a topical dermatitis agent. ... View more

Linda, There are some types of functions that work in different ways in SQL from the data step (they are called "aggregate" functions, like mean, sum, etc.). For those, look at the SQL reference manual. Your question is sufficiently naive that you may want to do some background reading. There are (at least) two Books-By-Users that are available on the SAS/Publishing web site that address SAS and SQL. Prairie's might be a good place to start, followed by Lafler's book. For your specific example, the SQL might look something like PROC SQL; CREATE TABLE xyz AS SELECT * FROM mydata.abc WHERE Eff_Frm_Dt GE TODAY() ; QUIT; RUN; as the result-equivalent of DATA xyz; SET mydata.abc; IF Eff_Frm_Dt GE TODAY(); RUN; Doc Muhlbaier Duke ... View more

The needs of the consumer. Safety is about whether a drug or device hurts people. Efficacy is about whether it helps people. ... View more

Before buying Learning Edition or the other possible solutions that Cynthia mentioned, check with your SFU computer support people. It may be that you already have a license for EG, but IT is not distributing it with SAS 9.1.3 (Our IT does not include it in the standard student distribution -- it involves a separate install, 2 additional CDs, and is not (yet) Vista aware.). Doc Muhlbaier Duke ... View more

Another possibility not already mentioned. If you have the physical tapes and no access to a mainframe, you can purchase a mainframe tape drive and attach it to a PC (usually with a SCSI Adapter). Some of them come with software to convert from EBCDIC text to ASCII text. You may even find some software that will allow you to tell the record format so you can unscramble the packed decimal code. Depending on the number of tapes, it may be simpler (and cheaper) to have a contract service do the conversion. ... View more

Sorry, it's not clear at all. Perhaps if you could show an example of two data sets with what you are getting a what you want to get, we could help. If you are using a DATA step MERGE, you should read the that section of the reference manual carefully; the details are important. If you are using PROC SQL, read read the similar sections on the types of JOINS. The rules for combining aren't 100% identical. Doc Muhlbaier Duke ... View more

SASman, Database tables are not the same as SAS tables. SAS tables (aka SAS datasets) on DASD keep the number of observations in the header. PROC SQL can't get that from an Oracle or SQL/server table. Also, SQL is "stateless" in the sense that it does not necessarily read or keep the observations in sequential order (that's why you need to use the ORDER BY clause even if you want the output to be in the same order that you knew the input to be). SQL uses indices to address order. What is your source file type. If you are somehow filtering this through some of the older file systems, you may be running against it's limits. Also check your system options (run PROC OPTIONS) to ensure your SAS install does not have a global limit set. Doc Muhlbaier ... View more

biyectivo, In addition to the other posts, I would add that SAS has a Performance Consulting Group who can help you make the best of what you've got or help with ideas on a purchasing decision. They were very helpful to us, both for the diagnosis of the problems on our existing system (and tuning to improve performance) and in providing us the objective information to take to management to get new systems in place. Doc ... View more