If you use the SELECTION= option in the MODEL statement, and the PARAM=REF, PARAM=EFFECT, or PARAM=ORDINAL option in the CLASS statement, and if there are missing values in some variables that are not included in the final model, then some predicted probabilities produced by the P= or PREDPROBS= option in the OUTPUT statement may be incorrect. In some cases the LOGISTIC procedure will appear to hang. The PARAM=GLM parameterization can be used to avoid the problem.

Operating System and Release Information

Product Family	Product	System	SAS Release
			Reported	Fixed*
SAS System	SAS/STAT	64-bit Enabled Solaris	9.1 TS1M0	9.3 TS1M0
		HP-UX IPF	9.1 TS1M0	9.3 TS1M0
		64-bit Enabled HP-UX	9.1 TS1M0	9.3 TS1M0
		64-bit Enabled AIX	9.1 TS1M0	9.3 TS1M0
		Microsoft Windows XP Professional	9.1 TS1M0	9.3 TS1M0
		Microsoft Windows Server 2003 Standard Edition	9.1 TS1M0	9.3 TS1M0
		Microsoft Windows Server 2003 Enterprise Edition	9.1 TS1M0	9.3 TS1M0
		Microsoft Windows Server 2003 Datacenter Edition	9.1 TS1M0	9.3 TS1M0
		Microsoft Windows NT Workstation	9.1 TS1M0
		Microsoft Windows 2000 Professional	9.1 TS1M0
		Microsoft Windows 2000 Server	9.1 TS1M0
		Microsoft Windows 2000 Datacenter Server	9.1 TS1M0
		Microsoft Windows 2000 Advanced Server	9.1 TS1M0
		Microsoft Windows Server 2003 Enterprise 64-bit Edition	9.1 TS1M0	9.3 TS1M0
		Microsoft Windows Server 2003 Datacenter 64-bit Edition	9.1 TS1M0	9.3 TS1M0
		Microsoft® Windows® for 64-Bit Itanium-based Systems	9.1 TS1M0	9.3 TS1M0
		z/OS	9.1 TS1M0	9.3 TS1M0
		Linux	9.1 TS1M0	9.3 TS1M0
		OpenVMS Alpha	9.1 TS1M0	9.3 TS1M0
		Tru64 UNIX	9.1 TS1M0	9.3 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.

---

Type:	Problem Note
Priority:	high
Topic:	Analytics ==> Categorical Data Analysis Analytics ==> Regression SAS Reference ==> Procedures ==> LOGISTIC

Date Modified:	2012-10-26 15:28:28
Date Created:	2012-10-26 14:59:24