06-10-2016 11:56 AM
I have the code for estimating mean cost and estimating the incremental effectivness Ratio using censored data from the SAS book "Analysis of Observational Health Care Data".
However, the code was written using the SAS?IML software. Unfortunately, i can not post the code here but anyone familiar with the code or if anyone knows any other SAS program to do the same please let me know.
Can anyone help/provide what modifications can i do to use it with base SAS?
Any help is greatly appreciated.
06-10-2016 01:13 PM - edited 06-10-2016 01:23 PM
The chance of someone helping will increase if you tell us which file contains the code, or provide a page reference for the book.The code from all SAS books are available for free. See Analysis of Observational Health Care Data Using SAS
If an author chooses to use SAS/IML software, it often means that the computation is cumbersome or unweildy in Base SAS procedures and the DATA step. When you provide more information, someone will advise you .
06-10-2016 01:24 PM
I wasn't aware of that. Thank you for letting me know.
It is the code from chapter 16. The code is on page 372-377. I am not familiar with IML at all and do not have SAS product either.
Thank you very much for your help in advance.
06-10-2016 01:44 PM
It looks like that code contains about 350 limes of IML programming, including a module that calls PROC LIFEREG and a DATA step. I don't know whether it is possible to convert it to Base SAS or not, but it is not a small or quick task. I certainly don't have the time to take on such a project.
I'll let other people post their opinions, but I am not optimisitc that you will find someone who can convert this for free. If I had to guess, I'd estimate that a consultant would need 20 hours or more to try to convert/test/validate this code. It might actually be cheaper to license IM if this operation is essential and will be done many times.
Do you work at a big pharmaceutical company? A medical school? A university? Many of those institutions actually license SAS/IML already.