BookmarkSubscribeRSS Feed
bermur
Calcite | Level 5
Hi everyone,

I have SAS 9.2.
This is more of a math side question hope someone has an answer.
I'm working with acceptance sampling plans specifically double sampling by variables.
So I have nonlinear function binomial pdf and which is iterated and calculated according to a array of probabilities [0..1].
I would like to equalize (I mean not exact match, equivalent) the result of my binomial pdf to bivariate normal distribution to find my 3 decision variables which are used to calculate bivariate function. As an example:

do p=0.0001 to 0.9999 by 0.0001;
Paatt = probbnml(p,n,c);

Zp = probit(1-p);

Pa1interup=sqrt(n1)*(Zp-k1);

Pa2interinup=sqrt(n1)*(Zp-k2);
Pa2interindwn=sqrt(n1)*(Zp-k1);
Pa2interoutup=sqrt(n1+n2)*(Zp-k3);

PA1 = probnorm(sqrt(n1)*(Zp-k1));

PAa = probbnrm(sqrt(n1+n2)*(Zp-k3), sqrt(n1)*(Zp-k2), n1/(n1+n2));
PAb = probbnrm(sqrt(n1+n2)*(Zp-k3), sqrt(n1)*(Zp-k1), n1/(n1+n2));

PA2 = PAa - PAb;

PA = PA1 + PA2;
end;

I would like to make the PA equivalent to Pattr by changing k1,k2,k3 for the whole nonlinear function. So its more of a approximation of k values from the result of Pattr vector. Any information may be helpful.
Thanks
1 REPLY 1
This forum is for Operations Research and Mathematical Optimization.
Maybe more relevant information can be found in the Statistics forum at the following link:

http://support.sas.com/forums/forum.jspa?forumID=46

hackathon24-white-horiz.png

The 2025 SAS Hackathon Kicks Off on June 11!

Watch the live Hackathon Kickoff to get all the essential information about the SAS Hackathon—including how to join, how to participate, and expert tips for success.

YouTube LinkedIn

Discussion stats
  • 1 reply
  • 1295 views
  • 0 likes
  • 2 in conversation