topic Re: Proc life test in Statistical Procedures

WEIGHT statement in Proc lifetest

aboli — Fri, 12 Jul 2024 18:33:07 GMT

Is there any option which we can give to get the correct censored summary while using weighted statement in proc life test?

Re: Proc life test

ballardw — Fri, 12 Jul 2024 19:27:15 GMT

It usually helps to show the code you are currently using. That way we can tell if any of your current options are conflicting with what you want to create. Copy you code and paste into a text or code box opened with the </> or "running man" icons above the message window.

For example, having your code would tell me immediately if this from the online documentation of the Weight statement for Proc Lifetest applies:

You can use the WEIGHT statement only for the Kaplan-Meier curves. Other methods of estimating the survival curves using weights are not available.

Re: Proc life test

data_null__ — Fri, 12 Jul 2024 17:24:08 GMT

Perhaps you are confusing the WEIGHT statement with the FREQ statement.

The FREQ statement identifies a variable that contains the frequency of occurrence of each observation.

Re: Proc life test

aboli — Sat, 13 Jul 2024 11:52:40 GMT

I am using below code. Output which got from below code sounds correct except censored summary output

ods output CensoredSummary = CENSUM
Quartiles=QUARTS
means = _mean
ProductLimitEstimates = KMEST1;

proc lifetest data=tte_os /*timelist= 3 , 6, 9, 12, 15, 21*/
outsurv=surv2 conftype=loglog /*reduceout*/ ;
time aval*CNSR(1);
strata trtn;
weight imt;
run;

Re: Proc life test

Season — Sat, 17 Aug 2024 12:13:57 GMT

I cannot see anything wrong with this code. What is your concern of the output?