TomKari Tracker

Re: Possible bug in proc sgpie

TomKari — Fri, 05 Jun 2026 17:33:03 GMT

Yes, I think that's the only solution. I asked for a clarification from Tech Support, it'll be interesting to see what they say.

Re: Possible bug in proc sgpie

TomKari — Thu, 04 Jun 2026 21:49:43 GMT

When I run it I'm still only getting one slice for v=5, size 1,000, and one slice for v=20, size 1. If I dial the count for v=5 back to 50, all of the slices show up. I want them to appear when v=1,000.

data pie1;
	do i=1 to 50;
		v=5;
		output;
	end;
	v=10;
	output;
	v=15;
	output;
	v=20;
	output;
run;

ods listing gpath='/home/xxx/AHAGM/images';
ods graphics / imagename="pie1" imagefmt=png;
title "pie1";

proc sgpie data=pie1;
	pie v / otherpercent=0;
run;

Re: Possible bug in proc sgpie

TomKari — Thu, 04 Jun 2026 00:07:29 GMT

No, what I'm trying to get is a slice for every value of v...so it should be count 1000 for 5, and count 1 for 10, 15, and 20. Instead, I'm only getting a count of 1 for 20.

Possible bug in proc sgpie

TomKari — Wed, 03 Jun 2026 16:48:23 GMT

Hi. When I run the following code...

data pie1;
	do i=1 to 1000;
		v=5;
		output;
	end;
	v=10;
	output;
	v=15;
	output;
	v=20;
	output;
run;

ods listing gpath='/home/xxx/AHAGM/images';
ods graphics / imagename="pie1" imagefmt=png;
title "pie1";

proc sgpie data=pie1;
	pie v / otherpercent=0;
run;

I get this result. My understanding is that using "otherpercent=0" should cause all slices to appear and be labelled.

Re: Using variable values in proc tabulate titles and headings

TomKari — Fri, 11 Jul 2025 23:37:43 GMT

Thank you all! It was the "#byval1" feature that I was fishing for. I knew it existed, but I've never used it and I simply couldn't put together a search string that found it. All the other great code that you guys have provided is a bonus!

Tom

Using variable values in proc tabulate titles and headings

TomKari — Thu, 10 Jul 2025 22:22:23 GMT

Hi, all

I have a fairly simple proc tabulate. Here's some code that emulates it (note the first data step is just to massage the data to somewhat closer to mine).

data pricedata;
	set sashelp.pricedata;

	if year(date)=1998;
	keep date productline productname sale price;
run;

title1 table of price data;

proc tabulate data=pricedata;
	class date productline productname;
	var sale price;
	table date, productname, productline * (sale*sum price*sum);
run;

It results in a table looks like this, with one page per month:

I would like to use the value of date in some of the headings. On the page title, instead of "table of price data" I would like to see "table of price data for jan98" on the first page, feb98 on the second, etc. Again, instead of "Name of product line" and "Product Name", I would like to see "Name of product line for jan98" and "Product Name for jan98"

All assistance much appreciated!
Tom

Re: How do I get a deplyment up and running.

TomKari — Tue, 04 Feb 2025 17:47:56 GMT

I tried following these instructions, and received the message shown in the text box. Not sure what to do next, please advise.

Tom

Re: BASUG virtual webinar by Art Carpenter

TomKari — Tue, 13 Feb 2024 20:49:09 GMT

I can't recommend this talk highly enough! Listening to Art wander through the decades is pure gold!!

Re: Background Submit in SAS On Demand for Academics

TomKari — Mon, 08 Jan 2024 19:52:28 GMT

I queried SAS Tech Support and they confirmed that indeed, background submission is not available in SAS On Demand.

Background Submit in SAS On Demand for Academics

TomKari — Fri, 05 Jan 2024 21:29:01 GMT

I am trying the following in SAS On Demand for Academics

"You can run one or more saved SAS programs in the background and continue to use SAS Studio while the background jobs execute. To run a program as a background job, right-click the program in the Server Files and Folders section of the navigation pane, and then select Background Submit from the pop-up menu."

But when I right-click my program, I only see the following:

How can I run a program in background?

Tom

Re: Loop macro 100 times

TomKari — Wed, 13 Sep 2023 17:38:01 GMT

Just to include my $.02 worth, i) I think it's great that Reeza is showing an alternate way of doing it, but ii) I absolutely agree with Paige that less-experienced developers should always use a PROC, if possible.

Tom

Re: Condition on Inner Join and the rest.

TomKari — Wed, 17 May 2023 03:08:37 GMT

SQL can be a little weird, especially when you're using something like EG, which tries to give you a user-friendly interface. It might be easier for you to do some experimenting, and see what happens. I suggest you try these as t1 and t2, and see what happens. Feel free to post questions. It'll give everybody a basis for discussion.

T!:

Recnum1	CIN
101	a
102	(null)
103	a
104	c
105	e
106	(null)
107	f
108	f
109	g

T2:

Recnum2	CIN
201	(null)
202	a
203	a
204	b
205	b
206	c
207	c
208	(null)
209	g

Re: Condition on Inner Join and the rest.

TomKari — Wed, 17 May 2023 02:42:35 GMT

It's a good question, but I think you've got yourself into a bit of a muddle. I explain a SQL join as "First, all of the records are matched with all of the records (a cartesian join)." "Second, based on your selection criteria, any records that don't match the criteria are removed". So, suppose t1.CIN values are a, c, e and t2.CIN values are b, c, e. After step one, you'll have 9 records with keys a, b; a, c; a, e; c, b; c, c; c, e; e, b; e, c; e, e. Now, step 2, follow the selection criteria. This will remove records c, c and e, e leaving you with 7 records.

Tom

Re: SAS Bowl XXXI, SAS Support Communities

TomKari — Thu, 11 May 2023 16:13:26 GMT

Something went wrong!?! I could see the game panel, but all of the interaction was missing! So sorry I missed you guys, hopefully the tech gremlins will have departed by the next one.

Tom

Re: General SAS Question

TomKari — Mon, 01 May 2023 20:44:36 GMT

In addition to Cynthia's response, which is absolutely correct, SAS has a number of tools that give an easier "look and feel" to meeting both data processing and statistical analysis requirements. Two very popular ones are SAS Enterprise Guide and SAS Studio. An interesting feature of both of them is that they create SAS code in the background, based on which options you select in the tool, and you can then use this code to learn SAS.

I suggest you ask your local SAS resources which products you have licensed, and let us know. We can then make some suggestions.

Tom

Re: SAS Bowl XXXI, SAS Support Communities

TomKari — Sun, 16 Apr 2023 16:56:52 GMT

What a great idea! It's been so long since I've been to a live SAS event, I have NO CLUE where my SAS swag is...😟

Re: Extract data from ASC file

TomKari — Fri, 03 Mar 2023 21:07:28 GMT

You need to provide a few details. What does your file look like? When you say ASC, do you mean an ascii text file? Can you include a couple of lines of the file?

Tom

Re: How to calculate monthly hospital visits and mortality rates in large clinical data?

TomKari — Mon, 19 Dec 2022 22:06:24 GMT

Okay, first some questions that will make it easier to provide advice:

1. You say a "large" clinical data set. Roughly how many records? Is it a flat file, a SAS dataset, or in a database?

2. I suggest you post a sample of a couple of dozen records. They can be fake, or have bits redacted, as long as we can see the fields that are important. Ideally, use this post to create some data for us to experiment with:

https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-data-AKA-generate/ta-p/258712

Again, don't post anything confidential.

Tom

Re: Scheduling Enterprise Guide via the management console

TomKari — Thu, 01 Dec 2022 21:32:10 GMT

Here's a post I made a while ago, and I still think it's valid.

Tom

==============================================

I recommend you use O/S scheduling (cron, scripts on Linux) capabilities or one of the SAS server-side products to schedule production jobs.

Enterprise Guide is a great tool, but I don't feel it's the best for scheduling production jobs on a remote server, as there are too many moving pieces (your machine, the network, and finally your SAS server). If any of these is having issues, it can interfere with your job executing correctly.

Use EG to develop and test your code, and for exploratory data analysis. Once you have your SAS programs going the way they should be, put them on the server and schedule them there.

If you search this site, you'll find a lot of posts that recommend this, in response to people who are experiencing problems using the EG scheduling facilities.

Re: Is there a reliable way to import calendar dates from excel to SAS in the EPG?

TomKari — Thu, 01 Dec 2022 21:22:02 GMT

I've used ddmmmyyyy for a number of years, so today would be 01Dec2022. I've never had a problem with it, but I haven't tested it exhaustively either.