topic Re: SAS procedure for Posterior probability in Statistical Procedures

SAS procedure for Posterior probability

manoj_pandey — Wed, 13 Jul 2022 14:07:04 GMT

Hi All,

I need to calculate the posterier probability for the objective response rate>=40 % based on the observed number of responders. I was looking for the whihc SAS procedure could be used but find the any example for the it. so could you please guide me for the same.

Thank you

Re: SAS procedure for Posterior probability

sbxkoenk — Wed, 13 Jul 2022 15:42:58 GMT

Hello,

Sorry. That question is not clear enough. Can you elaborate?
SAS has 100+ procedures that calculate posterior probabilities (adjusting for, taking into account or ignoring prior probabilities).

BR,

Koen

Re: SAS procedure for Posterior probability

manoj_pandey — Thu, 14 Jul 2022 01:48:08 GMT

Below are the details. Please let me know if you need more info.

There are 200 subject in study with 100 and 100 treated patients in TRT1 and TRT2, respectively, Objective response rate (ORR) can be estimated with a
maximum standard error of 0.041 and 0.071, respectively. Assuming a beta (0.5, 0.5) prior:

•TRT1: if 44 responders (out of 100 patients, ORR of 44%) are observed, the posterior probability of a true ORR >=40% will be >=0.80 (0.841).
•TRT2: if 46 responders (out of 100 patients, ORR of 46%) are observed, the posterior probability of a true ORR >=40% will be >=0.80 (0.807).

Re: SAS procedure for Posterior probability

sbxkoenk — Thu, 14 Jul 2022 10:22:48 GMT

Hello,

Go to https://www.lexjansen.com/

(all SAS papers from all major SAS conferences)

and enter as search terms (with quotes !!) : "Objective response rate"

Several papers are found with a.o. PROC FREQ for ORR.

Cheers,

Koen