BookmarkSubscribeRSS Feed
drhwint
Calcite | Level 5

I'm trying to analyze claim data using competing risk model.

 

Unfortunately, SAS Enterprise Guide does not support Gray's test using PROC LIFETEST. Maybe it contains SAS/STAT 13.x.

I knew there was an %CIF macro for Gray's test, it would not be helpful because I should consider the IPTW.

Is there any good way to get P value of Gray's test in this situation? 

 

Thanks. 

4 REPLIES 4
ChrisHemedinger
Community Manager

Moved this to the Stats Procs board.

 

Your version of SAS/STAT is independent of SAS Enterprise Guide -- the version of SAS/STAT depends on the SAS environment you connect to.  You can use proc product_status; run; to see the versions of software in your SAS environment.

Check out SAS Innovate on-demand content! Watch the main stage sessions, keynotes, and over 20 technical breakout sessions!
drhwint
Calcite | Level 5

Thank you for answering. 

 

When I checked with the code you sent, the SAS/STAT version was confirmed as 13.2.

SAS/GRAPH is identified as 9.4_M2.

 

Thanks. 

Tom
Super User Tom
Super User

Did you try just writing a code block to call the macro you mentioned?

 

Not sure what version of SAS/STAT you are running on the SAS server you are using with EG, but on mine here is the header for the autocall macro CIF.  Looks like it has existed since version 9.2 of SAS which was a long time ago.  Do you have IML and GRAPH licensed?

/*****************************************************************/
 /*          S A S   A U T O C A L L   L I B R A R Y              */
 /*                                                               */
 /*   NAME : CIF                                                  */
 /*   TITLE: Estimate Cumulative Incidence Function and           */
 /*          Compare Cumulative Incidence Functions across Samples*/
 /* PRODUCT: STAT                                                 */
 /*  SYSTEM: ALL                                                  */
 /*    KEYS: survival data, failure probability, competing risks  */
 /*   PROCS:                                                      */
 /*    DATA:                                                      */
 /* SUPPORT: Guixian Lin                UPDATE:  Mar2012          */
 /*          Ying So                    UPDATE:  Jun2014          */
 /*     REF: Klein  and Moeschberger   (2003)                     */
 /*          Marubini and Valsecchi    (1995)                     */
 /*          Gray                      (1988)                     */
 /*    MISC: The %CIF macro requires Base SAS and SAS/IML.        */
 /*          SAS/GRAPH is also required in SAS 9.2                */
 /*****************************************************************/

/*------------------------------------------------------------------
The %CIF macro computes the crude cumulative incidence function
estimates for homogeneous (no covariates) survival data whose
endpoints are subjected to competing risks (Klein and
Moeschberger, 2003). Standard errors and pointwise confidence
limits are also computed. The estimated cumulative incidence curve
is plotted as a  step function using ODS graphics. You can also
display the pointwise confidence limits in the plot by specifying
the PLOTCL keyword in the OPTIONS= argument. If you do not want any
plots, you can suppress it by specifying the NOPLOT keyword in the
OPTIONS= argument.

If the data contains multiple samples identified by the GROUP= variable,
the %CIF macro computes a cumulative incidence curve for each group, and
the macro also compares cumulative incidence functions between groups
using the method of Gray (1988).

...

 

drhwint
Calcite | Level 5

SAS/IML is version 13.2. I think, it is difficult to use the macro you sent because I know that weights cannot be applied in this macro.

 

Thanks. 

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

Register now!

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
  • 4 replies
  • 767 views
  • 0 likes
  • 3 in conversation