PaigeMiller Tracker

Re: Calculate time by comparing conditions and dates vertically

PaigeMiller — Mon, 08 Jun 2026 21:27:32 GMT

@Quentin is on the right track. Use the RETAIN statement. Give it a try, show us what you have tried.

Also, it would be extremely helpful if you can show us the desired output for this particular patient. (And hint hint hint, you should ALWAYS show us the desired output in these complex situations, we shouldn't have to ask.)

Re: Unable to open xls file

PaigeMiller — Thu, 04 Jun 2026 11:01:32 GMT

@srinivasa_attot wrote:

Statement:
proc export data=sample outfile=/home/raju/filename.xls" dbms=xls
replace;
run;

Is this a typographical error? You need a double quote after outfile= to match the double-quote at the end of the file name.

Re: selecting marker types and colors in proc sgplot

PaigeMiller — Wed, 20 May 2026 13:32:45 GMT

scatter x=Date y=Result / jitter group=Site  groupdisplay=cluster markerattrs=(size=9) ;

If you mean you want one value of site on the y-axis and another site should be on the y2-axis, I don't think this is possible here with the GROUP option.

On the other hand, if you create two separate variables, lets say SITE1 and SITE2, then you could have two different SCATTER statements, without GROUP=, and you can specify which of SITE1 and SITE2 goes on the y-axis and which goes on the y2-axis. So you need to re-arrange your data to meet this requirement.

Please from now on, paste code into a code box (as I have done). You do this by clicking on the "little running man" icon and pasting the code into the window that appears. Thank you.

Re: A SAS Macro to Use ODS to Direct Graphics Output to a Specified Folder and Avoid the WARNING: GP

PaigeMiller — Sun, 17 May 2026 10:08:17 GMT

@Tom wrote:

You get that message when your code produces graphs that need to be embedded/linked in the HTML output.

Most often is happens when you have not set the GPATH location and it defaults to the current working directory, but that directory is readonly. Most often for SAS sessions launched for use by Enterprise Guide or SAS/Studio since those typically start with the current working directory some place inside the SAS installation directories which normal SAS users cannot write to.

Which is why I never see this error message, I always use GPATH when I create graphics.

Re: proc pls : saving the model

PaigeMiller — Sat, 16 May 2026 19:02:06 GMT

This is the same thing I linked to in my original post, which you said was wrong, you said you can't use the training data set and test data set in the same PROC PLS.

Re: Combine the variable values vertically

PaigeMiller — Sat, 16 May 2026 15:10:10 GMT

@NewUsrStat wrote:
Thank you very much for your help. The reason why I have to replicate rows is because the dataset contains other variables with different values and for the subsequent analysis I have to apply different types of filters.

Unfortunately, this does not answer my question about what analyses/plots/tables are you trying to create.

Re: Combine the variable values vertically

PaigeMiller — Sat, 16 May 2026 15:05:28 GMT

In my opinion, this is not a good idea, and it makes future programming difficult. See Maxim 19. Also, why do you need an output of three identical rows?

Maybe a different re-arrangement of the data would work better. Please tell us what analyses/plots/tables you would create once you have the data re-arranged as you want?

Re: A SAS Macro to Use ODS to Direct Graphics Output to a Specified Folder and Avoid the WARNING: GP

PaigeMiller — Sat, 16 May 2026 11:49:58 GMT

I got tired of receiving error messages from SAS telling me that a particular folder was not authorized to receive output, as in

"WARNING: GPATH or PATH is not a writable directory. It will be ignored.

ERROR: Cannot write image to SGprint. Please ensure that proper disk permissions are set.".

I have never received this error. What am I missing, can't you just direct the output via ODS HTML or ODS LISTING to a writable directory in your original SAS program (which is what I do)?

Re: proc pls : saving the model

PaigeMiller — Sat, 16 May 2026 10:59:38 GMT

@SASdevAnneMarie wrote:
Thank you, but precisely, in Proc PLS we cannot use the train and test sets together in Proc.

Yes, you can use the test and train data set together, in certain situations. Obviously you are not in that situation.

Re: proc pls : saving the model

PaigeMiller — Sat, 16 May 2026 10:55:26 GMT

@Ksharp links to my exact solution for this problem. You would use the test data set in PROC SCORE, rather than the training data set, as I did in that example. PROC SCORE gives the predicted values using the PLS model. The final step where I create data set named COMPARE is not needed for the case where you want to score a test data set. The model is stored in data set COEFFS and COEFFS_T.

Re: proc pls : saving the model

PaigeMiller — Fri, 15 May 2026 18:27:05 GMT

@Rick_SAS gives the explanation of how to predict a test data set here: https://blogs.sas.com/content/iml/2014/02/17/the-missing-value-trick-for-scoring-a-regression-model.html

Re: proc pls method pcr plots

PaigeMiller — Thu, 14 May 2026 16:43:49 GMT

Please show us the full code for PROC PLS. Paste the code into the window that appears when you click on the 'little running man' icon. (From now on, when a PROC isn't working right, please show us the full code for that PROC automatically, don't wait for us to ask)

Are there warnings or errors in the log?

Re: How do organize he data to use ANOVA

PaigeMiller — Fri, 08 May 2026 23:26:01 GMT

What "combined means" are you talking about? Your description doesn't include any numeric variables. Please explain further.

Re: Assign variable values to time intervals

PaigeMiller — Thu, 07 May 2026 12:29:18 GMT

I asked for clarification on 3 issues, you only explained the first one.

If you are using informat mmddyy. then row 7 and 8 don't make sense, it shows month number equal to 30. (PS: advice is always to test your code first, make sure it works and does what you want it to do before providing it to us; you will get better and faster answers if you provide us with debugged code)

Re: Assign variable values to time intervals

PaigeMiller — Thu, 07 May 2026 11:34:57 GMT

A couple of questions

In the first record of data set DB, it seems to me that 9 months have elapsed but you show this should have a value of 3M in variable FLAG. I don't understand this.
You don't make clear if there is a different way to handle things when REF_DATE is before DATE_EVENT or when REF_DATE is after DATE_EVENT.
In your rules describing separation of months into 3M, 6M, 9M, etc., a literal reading of these rules seems like 3 months would get a FLAG value of BOTH 3M and 6M.

Ignoring these issues, and guessing at the answers about how to handle things, this is easily handled by the INTCK function and a custom format. Naturally if I have guessed wrong about how you want to handle things, you definitely ought to try fixing this code by yourself.

proc format;
    value monthf 0-3='3M' 4-6='6M' 7-9='9M' 10-12='12M' other='.';
run;
data db1;
    set db;
    delta_months=intck('month',ref_date,date_event);
    format delta_months monthf.;
run;

Why use custom formats here? Because if you use custom formats, the 3M 6M 9M 12M sort properly. If you don't use custom formats, 12M sorts before 3M.

Also, thank you for providing rules rather than (or in addition to) examples, and thank you for providing example data in the desired format.

Re: Coloring of words in SAS Viya/SAS Studio

PaigeMiller — Wed, 06 May 2026 14:59:41 GMT

@joeFurbee wrote:

@PaigeMiller,

I don't have an answer for why this occurs, but you may wan to fill out a Product Suggestion Form here in the Communities. They are reviewed on a regular basis.

At the moment, I don't want to change it. I want to understand how this coloring is helpful to me. If it turns out that it is not helpful, I will use this product suggestion form.

Coloring of words in SAS Viya/SAS Studio

PaigeMiller — Wed, 06 May 2026 14:21:09 GMT

In SAS Viya/SAS Studio, when I select a word by double-clicking on it, the word now has a blue background, and so does all the other occurrences of this word. What is the usefulness of coloring the background of all the occurrences of this word, rather than coloring only the one word I clicked on? How does coloring of all occurrences of this word help me perform any action? See example

Furthermore, if I get distracted for a second (or more), I don't remember which of these occurrences of this word I double-clicked on, they are almost identical shades of blue, how can I tell which one I double-clicked on? Is it possible to have a different background color for the one word I double-clicked on?

Re: Why I Still Trust This Community More Than Any AI Assistant

PaigeMiller — Wed, 06 May 2026 14:23:36 GMT

Thanks, @acordes , I will post a question in the VIYA forum in a minute. Update: here is my question

Re: Why I Still Trust This Community More Than Any AI Assistant

PaigeMiller — Wed, 06 May 2026 10:05:06 GMT

I too don't really trust any AI assistant when I need advice SAS advice that is more complicated than regurgitating the SAS help documentation.

I am new to SAS Viya, I wanted to do a relatively simple thing but couldn't find the answer. The simple thing I wanted to do was take some upper case code and turn it into lower case code. I know how to do this in SAS 9.4, but couldn't find the right option in Viya. So I asked here, I asked Google, and I asked Copilot.

I got the correct answer here in about 15 minutes. Google didn't understand the question. On the other hand, Copilot gave me the wrong answer, Copilot said it couldn't be done, but it could provide workarounds where it would write SAS code for me to perform the task programmatically. As I asked more questions, Copilot seemed to get irritated and aggressive in insisting that it was giving me the right answer and that there was no way to do this . An unpleasant process. Finally I told Copilot that I had found the answer and it gave in and admitted that it could be done. Out of curiosity, I asked Copilot again the next day, and it still told me it couldn't be done.

Re: proc pls : rmse values

PaigeMiller — Tue, 05 May 2026 18:40:09 GMT

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.