BookmarkSubscribeRSS Feed
☑ This topic is solved. Need further help from the community? Please sign in and ask a new question.
CraigSmith
Obsidian | Level 7

Hi all,

 

I am attempting to validate my logistic regression model in  a large dataset. 

 

However, it seems as if the max number of observations allowed for a calibration plot has been reached. Does anyone know how I could fix my code below. Have put error log below this.

 

Many thanks,

 

Craig 

 

CraigSmith_0-1685467571088.png

PROC LOGISTIC DATA=BIO3 PLOTS = ROC PLOTS(MAXPOINTS=NONE)
plots=calibration(CLM ShowObs);
MODEL HNC(EVENT='HNC') = PROB / ROCCI GOF OUTROC = ROC;
RUN;

 

 

 

NOTE: PROC LOGISTIC is modeling the probability that HNC='HNC'.
NOTE: Convergence criterion (GCONV=1E-8) satisfied.
NOTE: The number of observations of the LOESS plot (89450) exceeds the limit of 5000.
Specify the LOESSMAXOBS option of the ODS GRAPHICS statement to override the limit.
NOTE: The number of observations of the LOESS plot (89450) exceeds the limit of 5000.
Specify the LOESSMAXOBS option of the ODS GRAPHICS statement to override the limit.
NOTE: There were 89450 observations read from the data set WORK.BIO3.
NOTE: The data set WORK.ROC has 29701 observations and 8 variables.
NOTE: PROCEDURE LOGISTIC used (Total process time):
real time 10.92 seconds
cpu time 2.39 seconds

 

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
ballardw
Super User

SAS pretty much told you what to do with:

Specify the LOESSMAXOBS option of the ODS GRAPHICS statement to override the limit.

That means add, or modify your current, ODS Graphics with LOESSMAXOBS=89450, or larger.

 

You should have an ODS Graphics On somewhere in your code for Logistic to create the plots. So perhaps add this before the Proc Logistic.

ODS Graphics on / loessmaxobs=89450;

Your log does not show any errors. Those notes are information that you requested something that exceeds normal behavior for one option. Sometimes requesting large numbers of observations for such options doesn't really improve the quality of a graph or adds significant time for little added information.

View solution in original post

3 REPLIES 3
ballardw
Super User

SAS pretty much told you what to do with:

Specify the LOESSMAXOBS option of the ODS GRAPHICS statement to override the limit.

That means add, or modify your current, ODS Graphics with LOESSMAXOBS=89450, or larger.

 

You should have an ODS Graphics On somewhere in your code for Logistic to create the plots. So perhaps add this before the Proc Logistic.

ODS Graphics on / loessmaxobs=89450;

Your log does not show any errors. Those notes are information that you requested something that exceeds normal behavior for one option. Sometimes requesting large numbers of observations for such options doesn't really improve the quality of a graph or adds significant time for little added information.

CraigSmith
Obsidian | Level 7

Hey Ballard,

 

Thanks for the help - that did it.

 

It did run for an hour and a half but lacked memory due to the size. May need to consult with colleuages r.e a more powerful computer. 

 

Cheers 

 

🙂

AhmedAl_Attar
Rhodochrosite | Level 12

Depending on how/where you are running SAS software, there are invocation options that could allocate more memory to your SAS session!

  • Command Line: sas -memsize 6G
  • Windows Shortcut: Target: "[somePath\]sas.exe" "[somePath\]sasv9.cfg" -memsize 6G
  • Enterprise Guide connecting to Server ---> Talk to your SAS system administrator 🙂

 

sas-innovate-2024.png

Don't miss out on SAS Innovate - Register now for the FREE Livestream!

Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.

 

Register now!

What is Bayesian Analysis?

Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 3 replies
  • 562 views
  • 1 like
  • 3 in conversation