You can get the PLS equivalent of R-squared* for dependent variables, this is the value shown in Output 92.1.1 here: https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/statug/statug_pls_examples01.htm If you really need RMSE I think you could calculate the RMSE if you know the R-squared and total variation in Y, or from the residuals in Y.   * — if there is more than one dependent variable, this quantity is computed across all of the dependent variables. ... View more

Steps I've already tried: - Logging out and back in - Checking with my team if there are known permission restrictions   Did you contact your system administrator to request authorization? ... View more

@cbig wrote: Hello, the result of this code is giving me 0 observations.  The data does have duplicate rows per subjid so I expect there to be resulting rows.  The data is sorted correctly.  When I do "if first.subjid" I get the right number of rows.  It's just the "not" part that doesn't seem to be working. Can anyone see what I am doing wrong?  I'm so frustrated haha     proc sort data = ae_06c; by studyid site subjid; run;   data ae_06d; *has a 1; set ae_06c ; by studyid site subjid;   if not first.subjid ; run;   Hello, @cbig ! For this post and really for all your future posts, please provide us with some sample data that illustrates the problem. That would be very helpful and would likely result in you getting faster and more correct answers. Here are examples and instructions of how to provide us with sample data.   Without an example data set, I cannot figure out why your code doesn't work. It seems like it should work, however the main reason I can think of why it doesn't work is that your data isn't really what you said it was. So providing an example data set that illustrates the problem would straighten all of this out. Also, there are possibly important clues in the log. When your code is not behaving, please ALWAYS provide us with a copy of the log. Please copy the log for this DATA step (not just the possible ERRORs and WARNINGs but the entire log for this DATA step)and paste it into the Window that appears when you click on the </> icon. ... View more

So @Tom  has provided a very good solution. Let me try to explain why your original code doesn't work.   The %IF statement cannot access data set variable values. So you want to test to see if, for example, VAR2, has a value that is equal to 'Related'. %IF cannot do this, it has no idea what the values of VAR2 at any point in time. Data step tools, such as ARRAYs, the WHICHC function, and the good old IF statement, can do this.   So had the code been written with data step tools it might have worked. Please keep this in mind in the future when you try to write macros. ... View more

If I click on your link, I get taken to a rather unhelpful page, where there are apparently videos and articles for all different interfaces of SAS Viya, but none for SAS/Studio in SAS Viya. ... View more

I asked Copilot again today, and it insisted (several times) there was no way to do this in the SAS Viya interface. I suppose there are many things that AI can do well, but I have no confidence when I ask it SAS questions. ... View more

I would think the TEXT command in PROC SGPLOT would work, that's what it was designed to do (but I have not tried it). ... View more

Thanks, that works! Is there a web page that shows ALL keyboard shortcuts that exist by default in SAS Viya/SAS Studio? Adding: I tried asking Copilot after I asked the question here, however Copilot did not come up with the solution from @ChrisHemedinger , maybe if I try copilot again tomorrow it will find it. ... View more

Could you please provide more details? Which of the many SAS tools do you want to learn? What do you want to do with SAS?   Also please go back to your original post and change the title to a meaningful title that describes your problem in a few words. ... View more

All this variables (FruitWt, Color, Rind, Button, Wtloss) are numeric variables   It sounds like you are just stating this as a fact, rather than confirming this by running PROC CONTENTS. I have a rule here, when SAS says something is false and the human user says it is true, I believe SAS. (In my experience, SAS is right 99.99% of the time) So please check PROC CONTENTS. ... View more

The advice from @ballardw to represent Y and N with numeric values 1 and 0 should not be ignored. This applies to all SAS programming that you do now and in the future. If you are creating output (like a report or table) that you need to show others and Y and N is required in the output, then use a custom format to display 1 as Y and 0 as N.   However, I am still unable to understand the stated logic that takes me from the input data to the desired data. It simply is not clear.    Please try to write it again, in a new post in this thread (instead of correcting the original post) explaining the rules (in detail) you want to follow. As @ballardw would say, we need RULES and not EXAMPLES (although sometimes we need both). ... View more

@dsam wrote: I tried it, but month which is not numeric variable, order is messed up on x axis. Important concept: months and any calendar information in SAS data sets should be stored in NUMERIC variables. In fact, let me say that even more strongly: any calendar information in SAS data sets MUST be stored in NUMERIC variables. Character variables will not sort in the desired order for calendar information. ... View more

If I am understanding you properly, you will have to sort the data by the season and by the date of the race. Then you want to use the LAST. construct in SAS to in order to identify the last race of the season. Then it is easy to extract the information you want. See the documentation here, and the examples will be helpful.   Many people (including me) refuse to, or cannot, download attachments. The proper way to include a data set it to provide working SAS data step code (examples and instructions). ... View more

@dsam wrote: my code is below: I have x axis values like in months such as  D1 D28 2 3 4 6 12 18 24 36 but for one subject it is plotting month 18 after month 48 even though data is sorted properly by numeric variable.   title 'Percent Predicted FEV1 for Individual Subjects in Four 4D-710 Dose Cohorts'; proc sgplot data= combo; xaxis grid type= discrete discreteorder= data; series x = month y= RESTRESN/ group=subjid; scatter x=month y= RESTRESN/ group= subjid ; xaxis label= 'Month'; yaxis label= 'ppFEV1 (%)'; by cohort; run; When you have multiple XAXIS statements, the second one replaces whatever is specified in the first. ... View more