BookmarkSubscribeRSS Feed
salexgray
Calcite | Level 5

Hello everyone!

 

I'm relatively new to SAS and am currently running a Poisson regression with robust variance estimation. Within this code, I'm using the ESTIMATE statement to calculate the relative risks for each level of my IND2 variable (values from 0 to 5), and I am coming across this warning. Does anyone know how I can fix this? 

 

Also, what is the easiest way to graph these relative risks on one plot with the confidence intervals included? 

 

Thanks! 

 

proc genmod data=my.ddd descending;
class D_NARRID2 IND2(ref='0') R_ROLE2(ref='1') R_NARMS3(ref='1') R_NARED3(ref='1') R_AREA3(ref='1')/ param=ref;
model R_SS=IND2 R_UNROLE2 R_NARMS3 R_NARED3 R_AREA3/ dist=Poisson link=log type3;
repeated subject=D_NARRID2; * to get robust SE;
estimate '1 vs 0' IND2 -1 1 0 0 0 0 / exp;
estimate '2 vs 0' IND2 -1 0 1 0 0 0 / exp;
estimate '3 vs 0' IND2 -1 0 0 1 0 0 / exp;
estimate '4 vs 0' IND2 -1 0 0 0 1 0 / exp;
estimate '5 vs 0' IND2 -1 0 0 0 0 1 / exp;
run;
 
Error Log:
WARNING: More coefficients than levels specified for effect INDEX2. Some coefficients will be ignored.
WARNING: More coefficients than levels specified for effect INDEX2. Some coefficients will be ignored.
WARNING: More coefficients than levels specified for effect INDEX2. Some coefficients will be ignored.
WARNING: More coefficients than levels specified for effect INDEX2. Some coefficients will be ignored.
WARNING: More coefficients than levels specified for effect INDEX2. Some coefficients will be ignored.
3 REPLIES 3
PaigeMiller
Diamond | Level 26

Is this the log that goes with the code you show? The log talks about a variable INDEX2, which does not appear in the code you show.

--
Paige Miller
ballardw
Super User

When asking about warnings, errors or specific notes please copy from the LOG the entire procedure or data step code will all the warnings, notes and any errors. All the elements. Then paste into a code box opened on the forum with the </> icon.

The main message windows on the forum will reformat text and can make some diagnostics not appear correctly with the log entry.

The entire code clarifies any question about possible misspelled items when showing some of the code.

Ksharp
Super User
I think your parameterize is param=ref . But your 'estimate' is param=glm .
either change param=glm , or use 'ref' to 'estimate' .

Ready to join fellow brilliant minds for the SAS Hackathon?

Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.

Register today!
What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 3 replies
  • 1589 views
  • 1 like
  • 4 in conversation