BookmarkSubscribeRSS Feed
cherman
Calcite | Level 5

i've created a nice calibration plot for my logistic regression model using proc reg.  I have three calibration plots looking at the model calibration for three different subgroups. I am trying to display the graphs with the same vaxis so they are comparable.


here is my effort to control the Vaxis of my graph

ods graphics on;

proc reg data=crh;

plots=fitplot(stats=(adjrsq));

model x*y;

plot x*y/Vaxis=0 to 2.4 by 0.5;

run;

ods graphics off;

my issue is that my ODS output as a beautiful graph tieh 95% confidenec limits and prediction limits however the axis is not scaled.

I also get a second plot with the axis scaled as i specified. this plot is not as details as the ods output

My question is how do i get the ODS outputted graph to have the scaled axis as well??

regards

ch

3 REPLIES 3
Reeza
Super User

Unfortunately I think to customize those plots you'll have to modify the template.

You may be better off grabbing the output, prediction intervals etc and plotting using sgplot instead.

PS. Consider cross posting this in the graphics forum so one of the graph people can answer it Smiley Happy

cherman
Calcite | Level 5

thanks!

ballardw
Super User

Not critical but you will often get unexpected results in any of the axis controls when you use a value1 to value2 by value3 clause when value3 doesn't actually evently divide the interval. Your example might work better with 0 to 2.5 by 0.5 as the 2.4 isn't a "nice" multiple of 0.5.

hackathon24-white-horiz.png

The 2025 SAS Hackathon has begun!

It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.

Latest Updates

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

SAS Training: Just a Click Away

 Ready to level-up your skills? Choose your own adventure.

Browse our catalog!

Discussion stats
  • 3 replies
  • 1530 views
  • 0 likes
  • 3 in conversation