cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
MikeKim
Level IV

80% z-value : 1.3?

MikeKim_0-1699837785773.png

Well, above is the 415page of Pharmaceutical Quality by Desing Using JMP.
In above 90% at the last line is the Spec limit (LSL).

I am confusing the 1.3 looks like ensuring 90% Confidence. (1.282)

Can someone say that using 1.3 and this is 80% confidence?

 

I am sorry for the confusing question and many, but please help me.

1 ACCEPTED SOLUTION

Accepted Solutions
MRB3855
Super User

Re: 80% z-value : 1.3?

Hi @MikeKim : The 1.282 is correct for a one-sided 90% interval. And the z-value for a one-sided 90% bound is the same as the z-value for an 80% two-sided interval. So, while the page you reference is not worded very well (and exposes other issues that may be debated), a one-sided bound is appropriate for setting release limits (for a one-sided spec), and 1.282 is the right number for a 90% one-sided bound. For clarity, It should have been written referencing a one-sided bound (90%) rather than two-sided interval (80%), even though the z-value is the same. 

For more detail on this broad subject:

https://alfresco-static-files.s3.amazonaws.com/alfresco_images/pharma/2019/01/15/7b232584-09a9-4472-...

 

View solution in original post

1 REPLY 1
MRB3855
Super User

Re: 80% z-value : 1.3?

Hi @MikeKim : The 1.282 is correct for a one-sided 90% interval. And the z-value for a one-sided 90% bound is the same as the z-value for an 80% two-sided interval. So, while the page you reference is not worded very well (and exposes other issues that may be debated), a one-sided bound is appropriate for setting release limits (for a one-sided spec), and 1.282 is the right number for a 90% one-sided bound. For clarity, It should have been written referencing a one-sided bound (90%) rather than two-sided interval (80%), even though the z-value is the same. 

For more detail on this broad subject:

https://alfresco-static-files.s3.amazonaws.com/alfresco_images/pharma/2019/01/15/7b232584-09a9-4472-...