Samples & SAS Notes > Analytics
>
Survival Analysis
| Date | Title | Type | ||
|---|---|---|---|---|
| 11/20/09 |
37665 - The PROC PHREG BASELINE statement default METHOD= has changed in SAS 9.2 TS1M0 Beginning in SAS 9.2 TS1M0, the BASELINE statement METHOD=BRESLOW|CH|EMP|NELSON option is the default instead of METHOD=PL. http://support.sas.com/kb/37/665.html, 20KB |
Usage Note | ||
| 11/19/09 |
37830 - PROC PHREG uses additional memory beginning in SAS9.2 PROC PHREG allocates extra memory for CLASS variables and variable selection beginning in SAS9.2 TS1m0 http://support.sas.com/kb/37/830.html, 32KB |
Problem Note | ||
| 11/5/09 |
37621 - _TEMPORARY_ arrays in PROC PHREG are not initialized correctly _TEMPORARY_ array elements in PROC PHREG are set to zero for all observations regardless of their initial values. http://support.sas.com/kb/37/621.html, 20KB |
Problem Note | ||
| 11/5/09 |
37728 - Stratified trend tests in PROC LIFETEST may be incorrect for Peto, Modified Peto and Fleming tests Stratified trend tests may be incorrect for the Peto, ModPeto and Fleming tests unless the data are sorted by the GROUP= variable. http://support.sas.com/kb/37/728.html, 19KB |
Problem Note | ||
| 11/5/09 |
37577 - BASELINE statement CMF estimates ignore the OFFSET= option Cumulative Mean Function (CMF) estimates should equal the cumulative hazard estimates from the BASELINE statement CMF= and CUMHAZ= keywords. When the OFFSET= option is specified we no longer have equality between these ... http://support.sas.com/kb/37/577.html, 19KB |
Problem Note | ||
| 10/11/09 |
36285 - Memory problems might occur with a large number of BY groups If you are processing a large number of BY groups in PROC RELIABILITY then memory problems may occur. Sparse data (few failures or events) in some or all of the BY groups exacerbates this problem. http://support.sas.com/kb/36/285.html, 20KB |
Problem Note | ||
| 10/2/09 |
37170 - Variable name in ParameterEstimates table changed from 'Variable' to 'Parameter' The ParameterEstimates ODS table and data set for PROC LIFEREG in SAS 8.2, PROC TPHREG in SAS 9.1, and PROC PHREG in SAS 9.2 have changed. http://support.sas.com/kb/37/170.html, 20KB |
Usage Note | ||
| 9/8/09 |
36937 - BUS error may occur on UNIX operating systems with STRATA statement in LIFETEST A BUS error may occur on UNIX operating systems with specific configurations of the STRATA variable(s) http://support.sas.com/kb/36/937.html, 18KB |
Problem Note | ||
| 8/31/09 |
13679 - A binary or categorical covariate with no events in one level may cause monotone likelihood convergence problems resulting in huge standard errors. Categorical or binary covariates with no events in one level typically lead to convergence problems from a condition known as "monotone likelihood". The FIRTH option, first available in SAS9.2 is a potential remedy. http://support.sas.com/kb/13/679.html, 21KB |
Problem Note | ||
| 8/21/09 |
36453 - How to import and assess outside modeling techniques using SASŪ Enterprise Miner(tm) This video shows an example of how to import a non-supplied model into SAS Enterprise Miner and be able to assess it with other models. http://support.sas.com/kb/36/453.html, 21KB |
Usage Note | ||
