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
- /
- SAS/GRAPH and ODS Graphics
- /
- How to create a Parallel Coordinates Plot in SGPLO...

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
- Permalink
- Email to a Friend
- Report Inappropriate Content

01-23-2017 10:29 AM

I am hoping to create something similar to the Parallel Coordinates Plot from the Data Viz Catalogue (http://www.datavizcatalogue.com/methods/parallel_coordinates.html). The best solution I've found so far is from Prashant Hebbar's SGF 2012 paper (https://support.sas.com/resources/papers/proceedings12/267-2012.pdf) in which he converts each y variable to a percent and then uses a combination of vector, refline, and scatter statements to draw the multiple axes over the top of the series plot. While it works, this approach seems a little bit tedious. Is anybody aware of a more straightforward solution? Or, is this perhaps as simple as it gets?

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

Posted in reply to ShaneRosanbalm

01-23-2017 11:09 AM

If you are asking whether there is now a built-in statement for parallel coordinate plots, the answer is no.

However, there is a much easier way to standardize all numeric variables into [0,1]. You can use PROC STDIZE to transform all variables in a single call:

```
proc stdize data=sashelp.cars(drop=_CHARACTER_) method=range out=cars;
run;
/* view max/min for the transformed data */
proc means data=cars;
run;
```

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

Posted in reply to Rick_SAS

01-24-2017 09:23 AM

PROC STDIZE does tidy things up a bit. Thanks for the suggestion. That said, I will probably poke around a bit and see if I can't find a way to trick SGPLOT (or SGPANEL or GTL) into producing a slightly cleaner looking plot.