BookmarkSubscribeRSS Feed
kiwiepi
Calcite | Level 5

Hi everyone,

 

I wonder why the confidence intervals (variance estimates) are slightly different when I run the exact same proc surveylogistic syntax using SAS 9.4 and SAS 9.3.  For example:

 

proc surveylogistic data= abc varmethod= BRR;
weight WGT_FULL;
repweights bsw1-bsw500 / df =24;
class sex (ref= '1') / param = ref;
model SLP=sex;
run;

 

The confidence intervals are slightly wider when I used SAS 9.4.  Are there any differences between the two versions for this procedure?? I can't seem to find the information anywhere else.

 

Thanks!

2 REPLIES 2
ballardw
Super User

The most likely reason would be the refinement of algorithms in the coding of procedures between releases.

 

Secondary posibilities are a change from a 32-bit to 64-bit implementation.

Reeza
Super User

I would contact SAS Support with this, if you have concerns. It also depends on the differences, if their less than 0.0001 it's most likely precision issues more than anything else. 

SAS Innovate 2025: Call for Content

Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!

Submit your idea!

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
  • 2 replies
  • 1477 views
  • 0 likes
  • 3 in conversation