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
- /
- Create scatter plot with quadrants

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

05-29-2017 11:01 AM

Folks,

I would like to create a visual aid for my data in order to help me understand outliers.

What I would like to do is compare values obtained from a survey with values from other sources to see if there is a mismatch and where.

I was thinking of creating a scatter plot and divding it into four sectors.

My X-axis would be survey values and Y would be Other values.

Therefore the top left would be low survey value high other value, top right high survey value, high admin value and so on.

I've seen such an example for proc gplot bubble graph but I can't seem to replicate it for a scatter graph. Could anyone provide some assitance please?

Here is some sample data for people to look at.

```
data sample;
input id $ 2. survey other;
datalines;
01 15000 65000
02 6500 6500
03 8426 10000
04 1650 16805
05 630 1500
06 110055 56812
07 24568 15200
08 6000 7800
;
run;
```

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

Posted in reply to Sean_OConnor

05-29-2017 04:08 PM - edited 05-29-2017 04:13 PM

You don't mention how you intend to determine the cutoff values, but if you are looking at a regression context, you can look at the ROBUSTREG procedure, which automatically detects points of high-leverage and outliers. See the doc for the robust distance plot, which you can create as follows:

```
proc robustreg data=sample method=MM plots(only)=rdplot;
model other =survey;
run;
```

If you have your own method, you can use the REFLINE statement to add vertical and horizontal lines to your scatter plot:

```
proc sgplot data=sample;
scatter x=survey y=other;
refline 22400 / axis=x;
refline 20000 / axis=y;
run;
```

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

Posted in reply to Rick_SAS

05-30-2017 05:52 AM

Rick when I try to run the following command I get the following error.

```
ods graphics on;
22 proc robustreg data=varsofimport method=MM plots(only)=rdplot;
ERROR: Library COMMON does not exist.
ERROR: Library COMMON does not exist.
WARNING: Unable to load prototypes from 'common.funcs'.
23 model nat_totinc =nat_totinc_admin;
24 run;
```

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

Posted in reply to Sean_OConnor

05-30-2017 06:26 AM

1. What version of SAS?

2. What SAS interface? (Enterprise Guide? SAS Studio? SAS University Edition?)

3. Can you run the Getting Started example in the ROBUSTREG documentation?