turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- General Programming
- /
- npar1way DSCF error

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 03:40 PM

Hi,

I am running the following code is SAS 9.3:

proc npar1way data=set1 dscf wilcoxon;

class trial; var velocity;

run;

I am attempting to get Wilcoxon ranks for multiple comparisons based on the 'trial' variable. Trial has five levels. I am getting an error that SAS does not recognize the dscf option.

Syntax error, expecting one of the following: ;, AB, ALPHA, ANOVA, .......

I know that the dscf option wasn't introduced until 9.3, but I am using 9.3. Any ideas? I have also tried with adding in the 'exact wilcoxon' option.

Thanks

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 04:26 PM

I don't see that option listed in the docs...

were you looking for exact wilcoxon instead?

proc npar1way data=set1;

class trial; var velocity;

exact wilcoxon;

run;

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 05:27 PM

Here is the documentation:

Running the code:

proc npar1way data=set1;

class trial; var velocity;

exact wilcoxon;

run;

would give me a Kruskal-Wallis test to test the null hypothesis that v1=v2=v3=v4=v5, where vi is the mean of the ith trial. Instead, I want to test each pair of means for a difference using Wilcoxon rank scores. That is, the null hypotheses are v1=v2, v1=v3, v1=v4, etc. I know that are a lot of other ways of doing this but I am only interested in using the Wilcoxon ranks. Of course I could split the data up into 10 different sets with data from two different trials in each, and then apply the above code, but that would be time consuming (with that said I probably would have finished this hours ago if I had done it that way instead of spending all this time trying to figure the "easy" way of going about it).

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 05:59 PM

You're correct, its also in SAS/STAT 13.2 version.

When you start up SAS and it provides the version number what version of SAS/STAT does it mention?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 06:25 PM

When I start up SAS it mentions 9.3

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 07:18 PM

But no STAT version? I think you may just be a version or so behind. i.e. SAS 9.3 TS2M3 is the one that has SAS/STAT 12.1 etc...it usually will say in the log what version of SAS/STAT you have.

Or I'm totally off base...I'd do it the long way and contact tech support.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 07:21 PM

To see your version of SAS/STAT:

proc product_status;run;

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 07:58 PM

Aha! I was able to try running it on a different computer that had SAS 9.4 installed and it worked. The start-up message on the first computer was SAS/STAT 9.3_M1.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

10-01-2014 08:07 PM

Thank you for your help Reeza!