When the OBSMARGINS= or OM= option is specified in an LSMEANS, LSMESTIMATE, or SLICE statement, and the response variable is in the CLASS statement variable list in front of the class variables that also appear in the MODEL statement, the following modeling procedures generate incorrect results:

SAS/QC Software

• RELIABILITY

SAS/STAT Software

• GEE
• GENMOD
• LOGISTIC
• MIXED
• ORTHOREG
• PHREG
• PROBIT
• SURVEYLOGISTIC
• SURVEYREG

When the SAS/STAT postfitting procedure, PROC PLM, uses an item store that is generated by any of the above procedures or by the GLIMMIX procedure under the same conditions, and an OM-data-set is used, the least squares means results are also incorrect.

The LSMEANS, LSMESTIMATE, and SLICE statements with the OM= option generate incorrect results. The rest of the results are correct.

To circumvent the problem, specify the response variable after all of the CLASS statement variables that also appear in the MODEL statement. 

For example, the following syntax produces incorrect least squares means results:

Any permutation of the class variables that places the response variable after the model variables avoids the problem. Here is an example:

Click the Hot Fix tab in this note to access the hot fix for this issue.

Operating System and Release Information

Product Family	Product	System	Product Release		SAS Release
			Reported	Fixed*	Reported	Fixed*
SAS System	SAS/STAT	64-bit Enabled AIX	9.2		9.2 TS1M0
		64-bit Enabled HP-UX	9.2		9.2 TS1M0
		64-bit Enabled Solaris	9.2		9.2 TS1M0
		Linux for x64	9.2	2020.1.1	9.2 TS1M0	Viya
		Linux	9.2		9.2 TS1M0
		OpenVMS on HP Integrity	9.2		9.2 TS1M0
		Microsoft® Windows® for x64	9.2		9.2 TS1M0
		Windows Vista	9.2		9.2 TS1M0
		Microsoft Windows XP Professional	9.2		9.2 TS1M0
		Microsoft Windows Server 2003 Standard Edition	9.2		9.2 TS1M0
		Microsoft Windows Server 2003 Enterprise Edition	9.2		9.2 TS1M0
		Microsoft Windows Server 2003 Datacenter Edition	9.2		9.2 TS1M0
		Microsoft Windows XP 64-bit Edition	9.2		9.2 TS1M0
		Microsoft Windows Server 2003 Enterprise 64-bit Edition	9.2		9.2 TS1M0
		Microsoft Windows Server 2003 Datacenter 64-bit Edition	9.2		9.2 TS1M0
		Microsoft® Windows® for 64-Bit Itanium-based Systems	9.2		9.2 TS1M0
		HP-UX IPF	9.2		9.2 TS1M0
		z/OS	9.2		9.2 TS1M0

* For software releases that are not yet generally available, the Fixed Release is the software release in which the problem is planned to be fixed.

---

A fix for this issue for SAS/STAT 15.2 is available at:

https://tshf.sas.com/techsup/download/hotfix/HF2/I5J.html#66509

---

A fix for this issue for SAS/STAT 15.1 is available at:

https://tshf.sas.com/techsup/download/hotfix/HF2/E8U.html#66509

---

A fix for this issue for SAS/STAT 14.3 is available at:

https://tshf.sas.com/techsup/download/hotfix/HF2/C9Y.html#66509

---

If the response variable is in the CLASS statement variable list before the class variables that also appear in the MODEL statement, and an OM-data-set is used, least squares means results for several of the statistical procedures are incorrect.

Type:	Problem Note
Priority:	alert
Topic:	SAS Reference ==> Procedures ==> GEE SAS Reference ==> Procedures ==> GENMOD SAS Reference ==> Procedures ==> GLIMMIX SAS Reference ==> Procedures ==> LOGISTIC SAS Reference ==> Procedures ==> PLM SAS Reference ==> Procedures ==> PROBIT SAS Reference ==> Procedures ==> SURVEYLOGISTIC SAS Reference ==> Procedures ==> MIXED SAS Reference ==> Procedures ==> ORTHOREG SAS Reference ==> Procedures ==> PHREG SAS Reference ==> Procedures ==> RELIABILITY SAS Reference ==> Procedures ==> SURVEYREG Analytics ==> Regression Analytics ==> Reliability Analysis Analytics ==> Mixed Models Analytics ==> Longitudinal Analysis Analytics ==> Categorical Data Analysis Analytics ==> Analysis of Variance Analytics ==> Survey Sampling and Analysis Analytics ==> Survival Analysis

Date Modified:	2024-08-21 14:06:00
Date Created:	2020-08-20 16:12:33