topic Re: Proc Lifetest update impacting upper confidence interval in Statistical Procedures

Proc Lifetest update impacting upper confidence interval

sbrown999 — Mon, 18 Oct 2021 14:41:56 GMT

Hi, Since July I understand that there may have been an update to Proc lifetest which changes the upper confidence interval. I believe this may have been update on our system a couple of weeks ago.

http://support.sas.com/kb/64/617.html

So instead of getting an upper CI my CI is now "NE"

This is the data going into the procedure is below

proc lifetest data=in;

time pfsm*Cnsr(1) ;

run;

Old version CI (1.9,8.1)

new version CI (NE)

Point estimate is 8.11499

I'm not a Statistician so can anyone confirm that the new update is correct ? also is there a way to reproduce the old value?

Thanks

Stephen

Re: Proc Lifetest update impacting upper confidence interval

OsoGris — Mon, 18 Oct 2021 17:18:26 GMT

You have referenced the correct SAS Note regarding the quartile upper confidence limits (quartile UCL). There was a Hot Fix in the SAS Note that applied to SAS9.4 (TS1M6) = SAS/STAT 15.1 and is now a part of SAS9.4 (TS1M7) = SAS/STAT 15.2. Typically, after the fix is applied we see that some of the quartile UCL do not actually exist and are missing in the output. I think your current results are correct.

Re: Proc Lifetest update impacting upper confidence interval

sbrown999 — Tue, 19 Oct 2021 10:02:09 GMT

Thank you for your reply 😊