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
- /
- Analytics
- /
- Stat Procs
- /
- Fit plot with observations GLIMMIX

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
- Highlight
- Email to a Friend
- Report Inappropriate Content

04-13-2017 04:20 PM

I am using PROC GLIMMIX (SAS v9.3) to fit a mixed model in which I am most interested in the relationship between the outcome and a continuous covariate. I can get various graphical displays from PROC PLM, but I am having trouble finding anyway to plot the original observations on the effectplots. This is so easy in PROC GENMOD, I can't believe there isn't a way to acheive this with a model fit in GLIMMIX.

Thank you for any help!

Katy

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

04-13-2017 07:47 PM

PROC PLM doesn't have accesss to the original data, which is why it behaves differently. PLM only has access to the model.

In fact, the documentation for the OBS option in the EFFECTPLOT statement states "OBS<(obs-options)>

displays observations on the effect plots. An input data set is required; hence the OBS option is not available with PROC PLM."

Just use an OUTPUT statement to create a SAS data set that contains the predicted values. There are several predictions and scales (BLUP, ILINK,...) to choose from. Then you can sort by the covariate and use PROC SGPLOT to overlay the SERIES plot of the predictors on a SCATTER plot of the observations.